Professional Documents
Culture Documents
(Ronald A. McCauley) Corrosion of Ceramic and Comp
(Ronald A. McCauley) Corrosion of Ceramic and Comp
Editor
Philip A.Schweitzer, P.E.
Consultant
York, Pennsylvania
Ronald A.McCauley
Rutgers University
Piscataway, New Jersey, U.S.A.
10 9 8 7 6 5 4 3 2 1
v
Copyright © 2004 by Marcel Dekker, Inc.
vi Preface to the Second Edition
Ronald A.McCauley
vii
Copyright © 2004 by Marcel Dekker, Inc.
viii Preface to the First Edition
ACKNOWLEDGEMENTS
The author would like to thank the faculty and students of the
Department of Ceramic and Materials Engineering at Rutgers, The
State University of New Jersey, for many helpful and thoughtful
discussions during the preparation of this book and especially Drs.
John Wachtman and M. John Matthewson for reviewing a portion of
the manuscript and for their valuable suggestions.
The author would like to extend a very special thank you to Mr.
William Englert of PPG Industries, who first introduced the author to
the fascinating field of corrosion of ceramics.
Gratitude must also be extended to Mrs. Mary Guerin for her help
in preparation of the manuscript and to Paul Mort, Robert Sabia, John
Martin, and Ryan McCuiston for their help in preparing the figures.
The author would like to extend a very special thank you to his
wife, Eleanora, and his son, Matthew, for their understanding during
the many long hours required to complete this task.
Ronald A.McCauley
1 INTRODUCTION 1
2 FUNDAMENTALS 9
2.1 Introduction 9
2.2 Corrosion by Liquids 11
2.2.1 Introduction 11
2.2.2 Crystalline Materials 15
2.2.3 Glasses 47
2.3 Corrosion by Gas 56
2.3.1 Crystalline Materials 56
2.3.2 Vacuum 63
2.3.3 Glasses 64
ix
Copyright © 2004 by Marcel Dekker, Inc.
x Contents
Glossary 389
Epilog 393
1
Copyright © 2004 by Marcel Dekker, Inc.
2 Chapter 1
* University of Dayton Research Institute has the primary role. Georgia Institute
of Technology, North Carolina A&T State University, University of Missouri-
Rolla, along with Mechanical Test Instrumentation & Control are supplying
the information.
such as the fuel and raw materials saved during the shutdown.
Thus it should be obvious that the equation to determine the
exact costs of a particular repair is quite complex. The total
cost of such repairs can never be eliminated; however, it can
be greatly reduced by the proper selection of refractories and
the proper operation of the furnace. There are times when the
corrosion of the refractories goes unnoticed and failure occurs
prematurely. Since furnaces are insured against premature
failure, very large insurance claims have been filed. In addition
to the costs related to the refractories, construction, and lost
production are the costs related to additional cleanup due to
the failure and the costs of insurance adjusters and lawyers.
The cost of such a failure can exceed $20 million.
Environmental problems can also add to the total cost of
corrosion. For example, a trend toward the use of nonchrome-
containing refractories for furnace construction has been
ongoing for about the past 20 years. Used refractories have, in
the past, been disposed of by burying them in landfills. Chrome-
containing refractories have the potential of contaminating
groundwaters with hexavalent chrome, a carcinogen. If
chrome-containing refractories are used, upon disposal, they
must be hauled to toxic waste dumps with an added cost of
disposal. To eliminate this problem, some industries have been
leaning toward the use of other materials for the construction
of their furnaces. In some cases, the replacement material does
not last as long as the chrome-containing material, thus
shortening the time between repairs and adding to the cost.
The products of corrosion may enter the product being
manufactured and lower the quality of the product or decrease
the yields. Although this is a cost due to corrosion, it is one
that is extremely difficult to quantify. Although no accurate
numbers are available for the annual cost to industry for the
corrosion of ceramic materials, an estimate of $2 billion does
not seem unreasonable. Only through the intelligent selection
of ceramic materials can the cost of corrosion be minimized.
This intelligent selection of materials can be obtained only
through a thorough understanding of all the complexities of
ceramic materials and the effect that the environment has upon
them.
The corrosion of ceramic materials has not been as well
categorized as it has been for metals. Similar terms do, however,
appear in the literature. The more common types referred to
in the literature are diffusion corrosion, which is very similar
to concentration cell corrosion in metals, galvanic cell
corrosion, grain boundary corrosion, and stress corrosion. A
more common trend in ceramic materials is to group corrosion
under a more general mechanism, such as dissolution corrosion
(i.e., corrosion of a solid by a liquid). In this type of corrosion,
diffusion, galvanic cell, grain boundary, and stress corrosion
may all be present.
There are also many words used to describe corrosion, and
if one is looking for information on the subject of corrosion of
a particular material, a search including many of the following
words should be performed: dissolution, oxidation, reduction,
degradation, deterioration, instability, decomposition,
consumption, and erosion. Although erosion is technically not
the same as corrosion, being a physical effect rather than a
chemical one, erosion, in many cases, provides a means for
continued corrosion.
Corrosion of ceramic materials and its relationship to
various property degradation does not receive as wide a
recognition as it probably should at the various technical and
professional society meetings. For example, only about 2.5%
of the approximately 1500 presentations at the 2002 Annual
American Ceramic Society meeting were devoted to topics
related to corrosion. As can be seen from Table 1.1, the
percentage of presentations on corrosion has been between
4.5% and 2.5% during the decade of the 1990s. The data are
truly not sufficient to conclude that a trend exists. Although
the main topic of interest changes somewhat from year-to-
year, hazardous waste materials appear to remain a major
area of interest. Selecting a major topic is complicated by the
fact that areas often overlap, for example, general glass
durability vs. hazardous waste glass leaching. At meetings of
a
Includes posters.
b
Only presentations where corrosion was main topic.
c
Does not include hydration of cements and fabrication reactions.
d
HWM=hazardous waste materials.
2.1 INTRODUCTION
Corrosion of ceramics can take place by any one or a
combination of mechanisms. Various models have been
proposed to describe these mechanisms, several of which will
be discussed below. In general, the environment will attack a
ceramic, forming a reaction product. The reaction product may
be either solid, liquid, gas, or any combination of these. This
reaction product may remain attached to the ceramic or it may
be fugitive, in the case when gaseous species make up the
reaction product, or it may be a combination of both. When
9
Copyright © 2004 by Marcel Dekker, Inc.
10 Chapter 2
(2.2)
(2.3)
where:
g = acceleration due to gravity
∆ρ = (ρi-ρ∞)/ρ∞(ρi=saturated liquid density and
ρ∞ = original)
v = kinematic viscosity
x = distance from surface of liquid
Di = interface diffusion coefficient
C* = a concentration parameter
δ* = effective boundary layer thickness
R = solute radius
The exponential term was introduced as a correction for
cylindrical surfaces. Since experimental tests often involve
cylindrical specimens, these equations have been developed for
that geometry. In practical applications, the condition relating
to the corrosion of slabs is most predominant. However, if the
sample diameter is large compared to the boundary layer
thickness, the two geometries give almost identical results.
After a short induction period (which is of no consequence
in practical applications) in which molecular diffusion
predominates, the rate of corrosion becomes nearly
independent of time. As a surface corrodes, the interface, if
denser than the corroding medium, will be eroded away due
to free convection caused by density variation. Use of this
equation implies that one has at his disposal data relating to
the variation of density and viscosity with temperature. In cases
where these data are not available, the investigator will need
to determine them prior to any calculation of corrosion rates.
(2.4)
where:
jc = rate of corrosion
c = solubility of material in liquid
D = coefficient of binary diffusion
g = acceleration due to gravity
v = kinematic viscosity
L = distance from surface of liquid
ρ = relative variation of density
k = constant=0.482
This is essentially the same equation as that of Cooper and
Kingery, without the exponential term.
Many corrosive environments associated with ceramic
materials involve diffusion into the corroding medium, and
thus increased velocity of the medium increases corrosion. Thus,
if transport in the liquid were important, the corrosion rate
must be evaluated under forced convection conditions. In such
cases, the rate depended upon the velocity of forced convection:
(2.5)
The surface of the boulder shown in Fig. 2.5 was not horizontal
like the paver but sloped approximately 45°.
a
Solid, since melting point is 884°C.
Source: Ref. 2.21.
(2.7)
where:
Electrochemical Corrosion
Very few studies have been reported over the last 30 years;
however, much work was performed in the 1950s and 1960s
on what has been called galvanic corrosion of refractories by
glasses. Galvanic corrosion as defined by the physical chemist
must occur between two materials in contact with one another
and both must be in contact with the same electrolyte. Much
of what has been reported should more appropriately be called
electrochemical corrosion. One of the first reports of the
existence of an electrical potential between refractory and glass
was that of Le Clerc and Peyches [2.26] in 1953. The setup is
schematically represented in Fig. 2.6. In such a case, the molten
glass acts as the electrolyte and the platinum wire acts as a
reference electrode (i.e., standard oxygen electrode). The use
of platinum as a reference electrode requires that the
(2.8)
less negative free energy of formation for the oxide. The reaction
of aluminum metal with silica is a good example:
(2.10)
Since metals such as aluminum and magnesium have very large
negative free energies of oxide formation, determining what
to use as a container when melting these metals becomes a
serious problem.
Another mechanism for metal attack upon a ceramic is by
the formation of a new compound by a reaction of the type:
(2.11)
An example of this is the formation of spinel:
(2.12)
This reaction yields a lower free energy of reaction (∆G°1000=-
52 kcal/mol) than the simple redox reaction (∆G°1000=-28 kcal/
mol) shown below:
(2.13)
Thus a compound is formed that would appear to be a good
container for molten Al/Mg alloys. However, Lindsay et al.
[2.30] reported that reaction (2.13) was preferred to reaction
(2.12) due to the magnesium activity being sufficiently high to
form MgO and the reaction of A l 2O3 and MgO to form spinel
being kinetically slow.
Another possible reaction of a ceramic with a metal is that
of reduction to the metal and solution into the attacking molten
metal as shown below:
(2.14)
The metal that forms may be in the gaseous state depending
upon the environmental conditions, which may also be true
for the oxide that forms in reaction (2.13).
It has been found in the operation of commercial glass
(especially lead-containing glasses) furnaces that metals cause a
unique corrosion pattern on upward-facing horizontal surfaces
has four sp3 atomic hybrid orbitals (2s, 2px, 2py, and 2pz) that
are directed toward the corners of a tetrahedron. Each of the
hybrid orbitals contains two electrons. Two of these hybrid
orbitals react with a 1s hydrogen atomic orbital, leaving the
two remaining sp3 hybrid orbitals uncommitted (called the lone
pairs). This configuration thus produces an asymmetric electric
charge across the water molecule making water a polar
molecule. The small mass of the hydrogen allows the water
molecule to rotate so that the negative charge of the lone pairs
can interact with a positively charged ceramic surface or,
alternatively, form hydrogen bonds with a negatively charged
+ -
surface. The dissociation of water into ions of H and OH
most likely occurs as an excess or deficiency of protons on a
+ -
water molecule creating H3O and HO , maintaining electrical
neutrality. The relative concentration of these ions is the bases
for the pH scale of acidity and alkalinity.
A tremendous amount of information is available concerning
the leaching or dissolution of minerals, especially silicates, in
the soils literature, and anyone interested in corrosion in
aqueous systems should avail themselves of that literature. The
Jackson weathering sequence discussed by Marshall [2.33]
exhibits a trend that may be applicable, to some degree, to the
dissolution of silicate ceramics. The greater the degree of
bonding of the silica tetrahedra, the more difficult the
weathering. Some variation and overlap occurs due to the
specific chemistry for an individual mineral, with the minerals
containing alkalies and alkaline earths being less stable than
those containing alumina and the transition metals. This is
supported by Huang [2.34] who, in his studies of olivines,
pyroxenes, and amphiboles, reported that the relative stability
of these materials appeared, among other things, to be related
to the degree of polymerization of the tetrahedra, with more
highly polymerized materials being more stable. A related
phenomenon reported by Casey and Bunker [2.35] was that
minerals with a low density of cross-links tend to dissolve
congruently and rapidly, while minerals with a high density of
cross-links, such as the tektosilicates, dissolve incongruently
(2.16)
upon the interface reactions that take place. One must know
the environment and the ceramic thoroughly to understand
and explain any reactions or corrosion that takes place.
Although not an area that one would normally classify as
bioceramic, the whole area of the effects of mineral dusts upon
the human body are closely related to and actually very similar
to bioceramics. Anyone interested in this topic should consult
the review by Guthrie and Mossman [2.50]. In general, mineral
dusts that are inhaled may affect the tissues of the lungs. Since
lung fluids are undersaturated with respect to most chemical
species contained in minerals, these minerals will dissolve in
the lung. The health aspect of all this is a kinetics problem—
how long will it take to dissolve completely? Minerals that
dissolve within a matter of days are thought to have no health
effect, although this may not be true. Many studies have been
done to determine whether the health problem lies with the
chemistry of the minerals or with their morphology. Stanton
[2.51] and his colleagues have proposed a hypothesis that
mesothelioma (i.e., ability to form tumors) is caused by the
morphology of ceramic fibers. The general result was that
decreasing fiber diameter and increasing fiber length were
related to increased activity.
where:
P = hydration pressure
n = moles of water gained during hydration
R = gas constant
T = temperature
V = change in volume (volume of hydrate minus volume
of original salt)
Po = partial pressure of water vapor
Ph = vapor pressure of hydrated salt
An added aggravation can result from the freezing of water
with its associated expansion that can lead to cracking of the
building material. In addition, many marble monuments
develop a golden patina on the surface due to the selective
dissolution of the calcium contained in marbles with a small
content of iron [2.54]. This causes an enrichment of iron
compounds at the surface.
The porosity of limestone can increase if temperature
fluctuations (on the order of -15°C to +35°C) are present [2.55].
It has been reported that some marbles have exhibited as much
as 50% increase in porosity when subjected to temperature
fluctuations. The porosity increase is a function of the differential
thermal expansion of the calcite crystals present in limestone
and marbles. Calcite actually expands along the c-axis while
contracting in the perpendicular direction (αc=+26.5×10-6/°C vs.
αa=-5.7×10-6/°C). This expansion/contraction of the calcite
crystals causes them to separate forming small pores. Thus the
circulation of corrosive liquids increases causing increased
deterioration of the limestones.
In all the processes discussed above, it should be recognized
that water is an essential ingredient. If there were no water
present, there would be no dissolution or chemical reactions,
no transport of corrosive species, or no disruptive forces. The
chemical makeup of the water depends upon its origin: rainwater,
groundwater, or condensation (similar to rain). Condensation
is generally thought of as dew, but one must not forget that
frost and fog are also forms of condensation that may lead to
of limestone doubled for each pH=1 decrease. Thus one can see
that to be able to understand the problems of weathering of
ancient buildings, one must, in addition to being an engineer,
also be a historian and meteorologist.
Groundwater-related weathering implies that one knows a
little geology since dissolved species play a role in attacking
the sandstone. Local water tables and flooding history are
important aspects of the problem. To complicate the whole
process, a combination of the above mechanisms can take place.
The dissolution of limestones and marbles by atmospheric
corrosion takes place by one or a combination of the following
reactions [2.54]:
(2.19)
(2.20)
Since all atmospheres contain carbon dioxide, these reactions
will take place in the presence of rainwater. The temperature
and the CO2 partial pressure in the solution were reported by
Reddy et al. [2.60] also to affect the weathering rate. If any
sulfur is present in the atmosphere (generally from
industrialization that also increases the carbon dioxide
concentration), the following reactions take place:
(2.21)
(2.22)
(2.23)
These reactions are more severe than those shown in Eqs. (2.19)
and (2.20) above and the subsequent dissolution is more rapid.
(2.24)
This final reaction is one that leaves a white scum on the surface
of the limestone or marble through the evaporation of the water
and formation of gypsum. Vanadium oxides (a pollutant
FIGURE 2.8 Sulfuric acid dew point curves. (Reprinted with permission
of The Institute of Corrosion, United Kingdom. From Ref. 2.61.)
2.2.3 Glasses
Bulk Glasses
Probably the most abundant examples of glass corrosion are
those caused by a liquid. Release of toxic species (such as PbO
or radioactive waste) from various glass compositions has
reaction rates are quite often observed and are generally caused
by an excessively large exposed surface area due to microcracks
or generally rough surfaces. This excessive surface area can be
eliminated by proper cleaning procedures.
Jantzen [2.71] has used a thermodynamic approach to the
corrosion of glasses, especially applied to nuclear waste glass
leachability. The earlier work of Newton and Paul [2.72] on a
wide variety of glasses was expanded and then combined with
that of Pourbaix [2.73] and Garrels and Christ [2.74] to
describe the effects of natural aqueous environments. Using
thermodynamic hydration equations, Newton and Paul
predicted glass durability from composition. Jantzen showed
that the kinetic contribution was primarily a function of the
test conditions (SA/ V ratio,* time, and temperature). The major
assumptions in Jantzen’s approach were that the total free
energy of hydration of the glass was the sum of the free energies
of hydration of the components and that the glass structure
was a primary function of glass composition. The activity-pH
diagrams of Pourbaix provided the needed correlation between
free energy of hydration and ion concentration in solution.
Thus Jantzen was able to determine glass durability from glass
composition by use of a pH-adjusted free energy of hydration
term for several hundred compositions of nuclear waste glasses,
manmade glasses, and natural glasses. The more negative the
pH-adjusted free energy of hydration term, the less durable
the glass.
Species may be leached from a glass as a result of ion
exchange with protons from solution, or silica may be leached
as the siloxane bonds of the matrix are attacked by hydroxyl
ions from the solution. The former mechanism is predominant
at low pH, whereas the latter is predominant at high pH. Hench
and Clark [2.75] categorized leached glass surfaces into five
groups. These groupings are listed in Table 2.5. In Types I, II,
* SA/V ratio is the ratio of the surface area of the sample to volume of the
corroding liquid.
and III, the surface reaction layer that forms has a low solubility
if composed of metal hydrates or hydrated silicates. Quite often,
these layers are protective, essentially stopping further attack.
Many consider the pH of the solvent to be the most
important parameter that affects glass durability [2.76]. At
pH <5, ion exchange is the predominant mechanism, and at
pH >9, matrix dissolution is predominant. Between pH 5 and
9, the corrosion is a minimum. This is represented schematically
in Fig. 2.9. Thus dissolution is rapid when the metal-oxygen
bond is extensively coordinated to hydrogen or hydroxyl ions
and is a minimum under neutral conditions.
Since the driving force for dissolution of silicate glasses is
the hydrogen ion activity, the loss of hydrogen ions during
dissolution causes a continuous decrease in the dissolution rate.
The change of the pH of the solution can drift into the basic
region causing breakdown of the silica matrix with subsequent
increasing dissolution rates. To determine accurate rate
constants, the experimenter must hold the pH constant by use
of buffer solutions or an automatic titration system. Silica
solubility increases significantly above about pH=9.
In systems containing ions of variable valence, one must
consider the redox potential, Eh, of the system since the
solubility of ions is dependent upon their valence state. A
general rule of thumb is that higher valence states are more
soluble.
(2.25)
where:
W = weight loss
a = experimentally determined coefficient
b1 = experimentally determined coefficient
b2 = experimentally determined coefficient
φ = time
T = temperature
for describing the effects of time and temperature upon the acid
(10% HCl) corrosion of silicate glasses. This equation, since it
relates total multicomponent weight loss to time and temperature
assuming a uniform surface corrosion, does not take into account
the mechanism of dissolution, but instead determines the total
Fiber Glass
A discussion of glass would not be complete if some mention of
glass fibers were not made. The corrosion of fibers is inherently
greater than bulk glass simply because of the larger surface-to-
volume ratio. Since one of the major applications of fibers is as
a reinforcement to some other material, the main property of
interest is that of strength. Thus, any corrosion reactions that
would lower the strength are of interest. This effect is important
both when the fiber is being manufactured and after it has been
embedded in another material. For example, the strength of E-
glass (borosilicate) fibers in dry and humid environments was
studied by Thomas [2.81], with the observation that humid
environments lower strength. The mechanisms of
environmentally enhanced stress corrosion of glass fiber are
discussed in more detail in Chap. 8, page 360, Glassy Materials.
Wojnarovits [2.82] reported that multicomponent glass
fibers exhibited a variation in dissolution in acid and alkaline
environments due to the existence of a layered structure, each
having a different dissolution rate, with the core generally
having the highest rate. Single component fibers (i.e., silica)
did not show this layering effect and thus no variation in
dissolution rate.
Bioactive Glass
Bioactive glasses were first discovered by Hench in 1969. The
special chemistry of these glasses allowed them to bond to living
(2.29)
where:
y = film thickness
t = time
Ki = rate constant
Oxidation processes are generally more complex than the
simple mechanism of a single species diffusing through an oxide
layer. Preferential diffusion along grain boundaries can alter
the oxide layer growth substantially. Grain boundary diffusion
is a lower energy process than bulk diffusion and thus will be
more important at lower temperatures. Quite often, a higher
reaction rate will be observed at lower temperatures than
expected if one were to extrapolate from high-temperature
reaction rates. Thus the microstructure of the layer, especially
grain size, is particularly important. In addition, fully
stoichiometric reaction layers provide more resistance to
diffusion than anion- and/or cation-deficient layers, which
provide easy paths for diffusion.
Readey [2.91] has listed the possible steps that might be
rate-controlling in the kinetics of gas-solid reactions. These
are given below:
1. Diffusion of the gas to the solid
2. Adsorption of the gas molecule onto the solid surface
3. Surface diffusion of the adsorbed gas
4. Decomposition of reactants at surface-specific sites
5. Reaction at the surface
6. Removal of products from reaction site
7. Surface diffusion of products
8. Desorption of gas molecules from the surface
9. Diffusion away from solid
Any one of these may control the rate of corrosion.
(2.33)
One can then calculate the partial pressure ratio required for
the oxide or nitride to remain stable at any temperature of
interest. For example, the oxidation of silicon nitride to silica
at 1800 K yields a partial pressure ratio of nitrogen to oxygen
of about 107. Thus very high nitrogen pressures are required
to stabilize the nitride. Anytime the permeability of the product
gas through the reaction layer is less than that of the reactant
gas, the product gas pressure can build at the interface to very
high levels with the result being bubbles and/or cracks in the
reaction interface layer. This subsequently leads to continued
reaction.
The reduction of oxide ceramics at various partial pressures
of oxygen may also be of interest and can be obtained from
the examination of Ellingham plots of ∆G°=-RT In pO2 vs.
temperature (see Fig. 2.14 in Sec. 2.7.2). If one is interested in
the reduction of a binary compound, such as mullite, the
presence of a second more stable oxide that forms the
(2.36)
where:
Z = moles of gas that arrive at surface in unit time and
over unit area
P = partial pressure of reactant gas
M = molecular weight of gas
R = gas constant
T = absolute temperature
Using P and M of the product gas, the rate of removal of gas
product can be calculated using the same equation. To
determine if service life was acceptable, these rates may be all
that would be needed. Actual observed rates of removal may
not agree with those calculated if some surface reaction must
take place to produce the species that vaporizes. The actual
difference between observed and calculated rates depends on
the activation energy of the surface reaction. If the gaseous
2.3.2 Vacuum
It is generally believed that all materials vaporize; however,
several modes of vaporization are possible. Some materials will
vaporize congruently to a gas of the same composition as the
solid, which is also called sublimation. Others will vaporize
incongruently to a gas and a different condensed phase. It is
also possible for more than one stable gas molecule to form.
Decomposition to the elements may also occur, which is called
direct vaporization. In multicomponent materials where the
various components exhibit greatly different heats of
vaporization, selective vaporization may occur.
The deterioration of ceramics in a vacuum in many cases is
the equilibration of the material with a low partial pressure of
oxygen. In such a case, a lower oxide of the metal may form
along with some oxygen represented by the following equation:
(2.38)
(2.39)
where:
ro = initial radius
r = radius at time t
K = geometrical constant (~2)
D = diffusion coefficient of gas through boundary layer
Vo = molar volume of evaporating species
P = equilibrium partial pressure of gas
R = gas constant
T = temperature
t = time
2.3.3 Glasses
The corrosion of glasses by atmospheric conditions, referred
to as weathering, is essentially attack by water vapor.
Weathering occurs by one of two mechanisms. In both types,
condensation occurs on the glass surface; however, in one type,
it evaporates, whereas in the other, it collects to the point where
it flows from the surface, carrying any reaction products with
it. The latter type is very similar to corrosion by aqueous
solutions. The former type is characterized by the formation
of soda-rich films, according to Tichane and Carrier [2.99].
This soda-rich film has been shown to react with atmospheric
gases such as CO2 to form Na2CO3, according to the work of
Simpson [2.100] and Tichane [2.101].
The electronics industry is one area where vapor attack of
glasses may be of importance. Sealing glasses and glass
envelopes have been developed that resist attack by alkali
vapors and mercury vapors. In their study of some CaO- and
Al 2O3-containing glasses, Burggraaf and van Velzen [2.102]
reported that alkali vapor attack increased greatly above a
temperature that coincided approximately with the
transformation range* (Tg) of the glass, indicating that one
should use a glass with the highest possible Tg.
In the manufacture of flat glass by the Pilkington or PPG
processes, glass is floated onto a bed of molten tin in a chamber
containing a reducing atmosphere (N 2+~10%H2). The
hydrogen present in the atmosphere above the glass can act
upon the top surface of the glass causing reduction of the most
reducible species present. All commercial flat glass contains
some iron and that present near the top surface is predominantly
in the reduced ferrous state. This is generally not a problem;
however, those glasses containing NiO can exhibit small metallic
droplets on the top surface that are cause for rejection. Based
upon Fig. 2.14, this should not occur if the pO2 is maintained
greater than 10-9 atm, assuming a maximum temperature no
greater than 1100°C. Johnston and Chelko [2.103] proposed
the mechanism of reduction of ions in glass by hydrogen diffusion
through the glass to the reducible ions that act as immobile
traps reacting with the hydrogen and stopping further diffusion.
where:
D = diffusion coefficient
Do = constant
Q = activation energy
R = gas constant
T = absolute temperature
The larger the value of Q, the activation energy, the more
strongly the diffusion coefficient depends upon temperature.
The diffusion in polycrystalline materials can be divided
into bulk diffusion, grain boundary diffusion, and surface
diffusion. Diffusion along grain boundaries is greater than bulk
diffusion because of the greater degree of disorder along grain
boundaries. Similarly, surface diffusion is greater than bulk
diffusion. When grain boundary diffusion predominates, the
log concentration decreases linearly with the distance from the
surface. When bulk diffusion predominates, however, the log
concentration of the diffusion species decreases with the square
of the distance from the surface. Thus by determining the
concentration gradient from the surface (at constant surface
concentration), one can determine which type of diffusion
predominates.
Since grain boundary diffusion is greater than bulk diffusion,
it would be expected that the activation energy for boundary
diffusion would be lower than that for bulk diffusion. The
boundary diffusion is more important at lower temperatures,
and bulk diffusion is more important at high temperatures.
Chemical reactions wholly within the solid state are less
abundant than those which involve a gas or liquid, owing
predominately to the limitation of reaction rates imposed by
slower material transport. The solid-solid contact of two
different bulk materials also imposes a limitation on the
(2.41)
that the pores exist in a graded array with the largest ones
toward the outside of the ceramic as shown in Fig. 2.10. A
quick glance at Fig. 2.10 should convince anyone that
Washburn’s assumptions are far from reality.
One of the more controversial aspects of this technique is
the discrepancy between intrusion and extrusion data, which
has been explained by contact angle hysteresis by Smithwick
and Fuller [2.107]. Conner et al. [2.108] have shown the
sensitivity of this technique to pore morphology. Moscou and
Lub [2.109] reported that the hysteresis stems from a
combination of both contact angle differences for intrusion
and extrusion and pore morphology.
Lapidus et al. [2.110] and Conner and Lane [2.111] have
compared computer simulations of mercury flow through a
pore space assumed to be a pore-throat network to actual
porosimetry data and found that the throats determine the
intrusion behavior and the pores determine extrusion behavior.
The reader is referred to any of several review papers for more
detailed information [109, 112, 113].
(2.42)
where the subscript 1 denotes the hot face. If the gas pressure is
essentially the same on both sides, gases will migrate up the
thermal gradient in an attempt to make the pressure on the
hot face higher. The rate of migration is inversely proportional
to the square root of the molecular weight of the gas. Pore size
will affect the migration since very fine pores create too great
a resistance to flow and very large pores allow ordinary flow
due to pressure differences. Thus at some intermediate pore
size, transpiration will occur. In ceramics with a large pore size
distribution, ordinary flow tends to equalize the pressures,
minimizing flow by transpiration. There are no known reports
in the literature indicating that thermal transpiration influences
corrosion of ceramics; however, it may suggest a means to
minimize the effects from corrosive ordinary flow. If sufficient
flow of the transpiring gas is present, dilution of the corrosive
gas at the hot face may lower the corrosion rate to an acceptable
level.
The manufacturers of flat glass by one of the float processes*
are well aware of the problems that thermal transpiration may
cause. Although not a corrosion process, defective glass has
been produced by gases transpiring up through the tin bath
bottom blocks, rising through the tin, and then causing an
indent in the bottom surface of the glass. In some cases, the
gas pressure has been sufficient to puncture completely through
FIGURE 2.11 Surface area increase (a) due to dislocations and defects
and (b) due to ledge effect in fibrils.
where:
∆P = pressure drop across interface
γ = surface energy of liquid
φ = contact angle of meniscus at wall of capillary
R = radius of capillary
In actual materials, the porosity is not a cylindrical cavity and
therefore one must use an effective radius that represents the
weighted average of the contributing porosity. In addition, Eq.
(2.43) is valid for only an empty capillary. As soon as any
liquid penetrates the capillary, the driving force, ∆P, decreases.
The relationship of the surface energies among the solid-
vapor interface, solid-liquid interface, and the liquid-vapor
interface is given by:
(2.44)
(2.47)
* The evaluation of a 3-D parameter by use of a 2-D array will always lead to
a variation in values. Obtaining a representative value requires the determination
of a large number of individual dihedral angles. The final average value is
never the same as the actual value, but something less.
a
Substitution for MgO in an 85% MgO-15% CMS composition.
Source: Ref. 2.115.
that the dihedral angle between like grains was smaller than
that between unlike grains, indicating that the penetration of
liquid between unlike grains should be less than between like
grains.
The nature of the bonding type of the solid being attacked
compared to that of the attacking medium often can give an
indication as to the extent of wetting that may take place. For
example, transition metal borides, carbides, and nitrides, which
contain some metallic bond character, are wet much better by
molten metals than are oxides, which have ionic bond character
[2.116]. Various impurities, especially oxygen, dissolved in the
molten metal can have a significant effect upon the interfacial
surface energies. For example, Messier [2.117] reported that
silicon wet silicon nitride at 1500°C in vacuum but did not
spread due to oxygen contamination. In most cases, it is the
nature of the grain boundary or secondary phases that is the
controlling factor.
Puyane and Trojer [2.118] examined the possibility of
altering the wettability of alumina by using additives to their
glass composition. They found that V2O5 and CeO2 additions
changed the surface tension of the glass in opposite directions,
with V2O5 decreasing it and CeO2 increasing it. They concluded
that the glass characteristics were more important than the
solid parameters in corrosion.
2.7 THERMODYNAMICS
The driving force for corrosion is the reduction in free energy
of the system. The reaction path is unimportant in
thermodynamics, only the initial and final states are of concern.
In practice, intermediate or metastable phases are often found
when equilibrium does not exist and/or the reaction kinetics
are very slow. In general, a reaction may occur if the free energy
of the reaction is negative. Although the sign of the enthalpy
(or heat) of reaction may be negative, it is not sufficient to
determine if the reaction will proceed. The spontaneity of a
reaction depends upon more than just the heat of reaction.
There are many endothermic reactions that are spontaneous.
To predict stability, therefore, one must consider the entropy.
Spontaneous, irreversible processes are ones where the entropy
of the universe increases. Reversible processes, on the other
hand, are those where the entropy of the universe does not
change. At low temperatures, exothermic reactions are likely
to be spontaneous because any decrease in entropy of the
mixture is more than balanced by a large increase in the entropy
of the thermal surroundings. At high temperatures, dissociative
reactions are likely to be spontaneous, despite generally being
endothermic, because any decrease in the thermal entropy of
the surroundings is more than balanced by an increase in the
entropy of the reacting mixture.
In the selection of materials, an engineer wishes to select
those materials that are thermodynamically stable in the
environment of service. Since this is a very difficult task,
knowledge of thermodynamics and kinetics is required so that
materials can be selected that have slow reaction rates and/or
where:
F = Helmholtz free energy
E = internal energy
From Eqs. (2.49) and (2.50), it is obvious that the importance
of the entropy term increases with temperature. The reactions
of concern involving ceramic materials are predominately those
at temperatures where the entropy term may have considerable
effect on the reactions. In particular, species with high entropy
values have a greater effect at higher temperatures.
Gibbs free energy is a more useful term in the case of solids
since the external pressure of a system is much easier to control
than the volume. The change in free energy is easy to calculate
at any temperature if the enthalpy and entropy are known.
(2.53)
where:
n = number of electrons involved
F = Faraday constant (96,490 C/mol)
E = standard cell potential (volts)
Tabulations of standard half-cell potentials (standard emf
series) are available and are more commonly called redox
potentials [2.127]. The use of the emf series for studies in
aqueous solutions has been established for a long time and has
now been extended to nonaqueous electrolytes such as molten
salt mixtures. According to Brenner [2.128], who reported
average errors of 32% between calorimetric and emf
measurements, the use of Eq. (2.54) is not accurate and it should
be modified as required for each galvanic cell evaluated.
Although industrial process gas streams are generally not
in thermodynamic equilibrium, their compositions are shifting
toward equilibrium at the high temperatures normally
encountered. Using equilibrated gas mixtures for laboratory
studies then is a basis for predicting corrosion but is not
necessarily accurate. Which reaction products form at solid/
gas interfaces can be predicted from free energy calculations
using the following equation:
(2.55)
(2.56)
(2.57)
(2.58)
(2.60)
rearranging:
(2.61)
or:
(2.62)
At constant temperature, if the partial pressure of CO2 over
CaCO3 is maintained at a value less than kp, all the CaCO3 is
converted to CaO. If the partial pressure of CO2 is maintained
greater than kp, then all the CaO will react to form CaCO3.
This type of equilibrium, involving pure solids, is different from
other chemical equilibria that would progress to a new
equilibrium position and not progress to completion.
An example, similar to the above description for Eq. (2.57),
for a reaction when both the reactants and products are all
solid phases was given by Luthra [2.130] for the reaction of
an alumina matrix with SiC reinforcement fibers. The following
equation depicts this reaction:
(2.63)
FIGURE 2.15 Ellingham type diagram for the Al–O system. (From
Ref. 2.134, reprinted with permission of The American Ceramic
Society, www.ceramics.org. Copyright © 1985. All rights reserved.)
2.8 KINETICS
It is normally expected that materials will corrode, and thus it
is important to know the kinetics of the reaction so that
predictions of service life can be made. Thus the most important
(2.68)
the mass of material reacted per unit area for a unit time. These
can easily be converted to the depth of penetration per unit time
by dividing by the density of the material as shown below:
(2.71)
where:
P = depth of penetration
M = mass loss
ρ = density
A = area of exposure
t = time of exposure
In using the above equation to calculate corrosion rates from
laboratory experiments, one must be very conscious of the total
surface area exposed to corrosion. This will include a
determination of the open porosity of the specimen. Many
investigators have attempted to compare corrosion resistance
of various materials incorrectly by omitting the porosity of their
samples. Omitting the porosity, although not giving a true
representation of the material’s corrosion, will give a reasonable
idea of the corrosion of the as-manufactured material.
2.9 DIFFUSION
When the transport of ions or molecules occurs in the absence
of bulk flow, it is called diffusion. Substances will spontaneously
diffuse toward the region of lower chemical potential. This
transport or flux of matter is represented by Fick’s first law
and is proportional to the concentration gradient. This is
represented by:
(2.72)
where:
Jix = flux of component i in the x direction
D = diffusion coefficient
ci = concentration of component i
(2.73)
(2.74)
where:
C(x,t) = concentration after time t
Co = initial concentration in the medium
(2.75)
(2.76)
(2.77)
REFERENCES
2.1. Phase Diagrams for Ceramists. Vol. I-XII, Am. Ceram. Soc.,
Westerville, Ohio.
2.2. Cooper, A.R. The use of phase diagrams in dissolution studies.
In Refractory Materials; Alper, A.M., Ed.; Academic Press:
New York, 1970; Vol. 6-III, 237–250.
2.3. Kramer, D.P.; Osborne, N.R. Effects of atmosphere and dew
point on the wetting characteristics of a glass-ceramic on two
nickel-based superalloys. In Ceramic Engineering and Science
Proceedings; Smothers, W.J., Ed.; Am. Ceram. Soc.
Westerville, Ohio, 1983; 4 (9–10), 740–750.
2.4. Noyes, A.A.; Whitney, W.R. Rate of solution of solid
materials in their own solutions. (Ger) Z.Physik. Chem.
1897, 23, 689–692.
2.5. Nernst, W. Theory of reaction velocities in heterogeneous
systems. (Ger) Z.Physik. Chem. 47, 52–55.
2.6. Berthoud, A. Formation of crystal faces. J.Chem. Phys. 10,
624–635, 1912.
2.7. Prandtl, L. NACE Tech. Memo. No. 452, 1928.
2.8. Levich, B.G. Physicochemical Hydrodynamics; Prentice Hall:
Englewood Cliffs, NJ, 1962.
2.9. Levich, E.G. Theory of concentration polarization. Discuss.
Faraday Soc. 1, 37–43, 1947.
2.10. Cooper, A.R., Jr.; Kingery, W.D. Dissolution in ceramic
systems: I, Molecular diffusion, natural convection, and
forced convection studies of sapphire dissolution in calcium
aluminum silicate. J. Am. Ceram. Soc. 1964, 47 (1), 37–3.
2.11. Samaddar, B.N.; Kingery, W.D.; Cooper, A.R., Jr. Dissolution
in ceramic systems: II, Dissolution of alumina, mullite,
One must learn by doing the thing; for though you think
you know it, you have no certainty until you try.
SOPHOCLES
3.1 INTRODUCTION
According to Weisser and Bange [3.1] it was Lavoisier who in
1770 first recorded the aqueous corrosion of a silicate glass
predominantly by use of an analytical balance. The analysis of
corrosion has been changing over the years with the greatest
changes probably taking place within the last 25 years. These
changes have been due mostly to the availability of
sophisticated computerized analytical tools. It has taken many
years for investigators to become familiar with the results
obtained and how to interpret them. In some cases, special
123
Copyright © 2004 by Marcel Dekker, Inc.
124 Chapter 3
lodged within the pores and cracks of the sample being prepared.
The final grinding media grit size should be no greater than 15
µm. It is best to clean samples after each grinding step in an
ultrasonic cleaner with the appropriate cleaning solution.
Surface roughness of solid samples is an item that is often
overlooked. Not only does a rough surface increase the area
exposed to corrosion, but it may also lead to problems with
some analytical techniques. For example, when the surface
roughness is on the order of the reaction layer thickness caused
by corrosion, errors will be present in the depth profiles
obtained by secondary ion mass spectroscopy (SIMS). In those
cases when surface analyses are planned, one should prepare
solid samples to at least a 5-µm finish.
Grinding and polishing of samples that contain a reaction
product surface layer should be done so that the reaction layer
is not damaged, or the interface obscured. If part of the sample
is metal, then polishing should be done in the direction ceramic
toward metal to eliminate smearing the metal over the ceramic.
If very thin reaction layers are present, one can prepare taper
sections to increase the area that is examined. Sample preparation
of composites presents some additional problems since materials
of very different characteristics will be presented at the surface
being polished. Chanat [3.4] of Buehler Ltd. has offered some
good advice for mounting, sectioning, and polishing. The most
effective method involves the use of diamond abrasives with
low nap cloths for FRCMC*. High nap cloths can induce
excessive relief at boundaries of different materials.
Many tips on how to prepare samples can be obtained by
reading the various technical journals published by the
manufacturers of consumable grinding and polishing supplies.
One particular article that offers some new ideas was that of
Damgaard and Geels [3.5]. They emphasized the importance
of polishing disk diameter and velocity, indicating that both
are directly proportional to material removal rates. Although
* One should also remember that a perfectly polished surface, although excellent
for reports, is not necessary to obtain sufficient data to solve a corrosion problem.
and
(3.2)
(3.3)
features and not just what one observes from the topographic
features. Although SEM can be performed on as-received or
rough surfaces, EDS is best performed on polished or flat
surfaces. The analysis by SEM/EDS in combination with XRD
and optical microscopy is a powerful tool in the evaluation of
corrosion. See Fig. 5.3, which shows optical, SEM/EDS, and
XRD data for the corrosion of a mullite refractory, and the
corresponding text for an example of the use of EDS in phase
identification.
Transmission Electron Microscopy
Transmission electron microscopy (TEM) can be used to
evaluate the corrosion effects upon grain boundary phases.
Transmission electron microscopy used in this way can be very
useful; however, it is a very time consuming method, and, quite
often, the samples are not representative due to their small
size (several millimeters or less) and the thinning process.
Transmission electron microscopy does not lend itself to the
observation of porous samples and thus is confined to
observation of dense regions of corroded samples.
* Helium is more penetrating than water and thus yields a smaller volume
determination. This is dependent upon the pore size distribution.
REFERENCES
3.1. Weisser, M.; Bange, K. Sophisticated methods available to
analyze glass corrosion. Glass Res. 2000, 9 (2), 16–17, 21.
3.2. Wachtman, J.B. Characterization of Materials; Butterworth-
Heinemann: Boston, 1993.
3.3. Brady, P.V.; House, W.A. Surface-controlled dissolution and
growth of minerals. In Physics and Chemistry of Mineral
Surfaces; Brady, P.V., Ed.; CRC Press: New York, 1996; 225–
305. Chp. 4
3.4. Chanat, S. Preparation techniques for analysis of fiber
reinforced ceramic matrix composites. In Ceramic Transaction,
Advances in Ceramic-Matrix Composites; Bansal, N.P., Ed.;
Am. Ceram. Soc.: Westerville, OH, 1993; Vol. 38, 603–615.
3.5. Damgaard, M.J.; Geels, K. High capacity materialographic
specimen preparation. Struers J.Materialogr. 2001; 7–11.
Structure 38.
3.6. Macchesney, J.B.; Rosenberg, P.E. The methods of phase
equilibria determination and their associated problems . In
Phase Diagrams: Materials Science and Technology; Alper,
A.M., Ed.; Refractory Materials; Margrave, J.L., Ed.; Academic
Press: New York, 1970; Vol. 6–1, 113–165. Chp. 3.
3.7. Eriksson, G. Thermodynamic studies of high temperature
equilibria. XII. SOLGASMIX, A computer program for
calculation of equilibrium compositions in multiphase
systems. Chem. Scr. 1975, 8, 100–103.
3.8. Cherry, R.J., Ed.; New Techniques of Optical Microscopy
and Microspectroscopy; Topics in Molecular and Structural
Biology; Neidle, S., Fuller, W., Series Eds.; CRC Press: Boca
Raton, FL, 1991.
3.9. Alexander, L.; Klug, H.P. Basic aspects of X-ray absorption.
Anal. Chem. 1948, 20, 886–889.
3.10. Chung, F.H. Quantitative interpretation of X-ray diffraction
patterns of mixtures: I. Matrix-flushing method for
quantitative multicomponent analysis. J. Appl. Cryst. 1974,
7, 519–525.
3.11. Dickson, M.J. The significance of texture parameters in phase
analysis by X-ray diffraction. J. Appl. Cryst. 1969, 2, 176–180.
3.12. Brime, C. The accuracy of X-ray diffraction methods for
determining mineral mixtures. Mineral. Mag. 1985, 49 (9),
531–538.
3.13. Schmidt, C.; Rickers, K. In-situ determination of mineral
solubilities in fluids using a hydrothermal diamond-anvil cell
and SR-XRF: Solubility of AgCl in water. Am. Mineral. 2003,
88 (2–3), 288–292.
When you can measure what you are speaking about and
express it in numbers you know something about it; but
when you cannot measure it, when you cannot express it
in numbers, your knowledge is of a meager and
unsatisfactory kind.
LORD KELVIN
4.1 INTRODUCTION
The American Society for Testing and Materials (ASTM) was
formed in 1898 through the efforts of Andrew Carnegie and the
chief chemist of the Pennsylvania Railroad, Charles Dudley,
who were both convinced that a solution was necessary to the
unexplainable differences of testing results that arose between
their laboratories. These early efforts were focused upon
143
Copyright © 2004 by Marcel Dekker, Inc.
144 Chapter 4
* The MCC was created in 1980 by the U.S. Department of Energy and is
operated for the DOE by the Pacific Northwest Laboratories of the Battelle
Memorial Institute in Richland, WA.
size. The samples are then dried at 23°C and a relative humidity
of 50%. The length change is recorded after 4, 7, 14, and 28
days and 8, 16, 32, and 64 weeks.
At least three 2-in. cubes per test medium and per test
temperature are immersed into approximately 150 mL of the
desired test liquid. The closed containers are placed into a
constant temperature oven or bath and then examined after 1,
7, 14, 28, 56, and 84 days. The samples are evaluated for
weight change and compressive strength change.
REFERENCES
4.1. Form and Style for ASTM Standards, 7th Ed. ASTM:
Philadelphia, March 1986.
4.2. Mendel, J.E. (compiler). Nuclear Waste Materials Handbook-
Waste Form Test Methods, Materials Characterization Center,
Pacific Northwest Laboratories, Richland, WA, U.S. DOE
Report DOE/TIC-11400, 1981.
165
Copyright © 2004 by Marcel Dekker, Inc.
166 Chapter 5
Alumina-Containing Materials
*
Tetragonal zirconia transforms to monoclinic zirconia athermally upon
cooling at about 1000°C accompanied by an expansion of about 5 vol.%.
face of the wall was present due only to gravity. The alteration
of the refractory due to corrosion occurred to a depth of about
25 mm. As can be seen from Fig. 5.3, the mullite has completely
converted to predominantly corundum and nepheline. Fig. 5.3a
is the X-ray diffraction (XRD) pattern supporting the presence
of only corundum and nepheline. The optical micrograph shown
in Fig. 5.3b indicated the presence of an additional phase. Upon
examination of elemental maps via scanning electron
microscopy (SEM)/energy dispersive spectroscopy (EDS) shown
in Fig. 5.3c, it was determined that the nepheline contained a
reasonable amount of dissolved calcium and that the crystalline
nepheline was embedded in a matrix of vitreous potassium-
titanium-silicate. The potassium diffused into the refractory from
glass batch impurities and the titanium was present in the
original refractory in minor amounts.
Zircon
The attack of zircon by soda-lime-silicate glasses is similar to
that of AZS materials in that an interface of zirconia crystals
embedded in a highly viscous glass is formed. The difference is
the lack of alumina, which keeps nepheline from forming and
the viscous glass is now a silicious glass as opposed to an
alumina-rich glass. Thomas and Brock [5.17] reported that as
the sodium content of the attacking glass decreased, the thickness
of the zirconia layer decreased. The attack of zircon by E-glass
(a borosilicate) that contains only about 0.5% Na2O exhibited
no observable alteration. Zircon has been successfully used in
contact with high-temperature lithium-alumino-silicate glasses.
A protective layer of zirconia crystals suspended in a very
viscous glass is formed by the leaching of silica from the zircon.
Because these glasses are melted at temperatures above 1700°C,
it is quite possible that the zircon dissociates into zirconia and
silica, with the silica then going into solution rather than the
Zirconia
Because of the polymorphic transitions associated with zirconia,
it has not been widely used as material for furnace linings except
when combined with other materials, such as the AZS refractories
described above, or when stabilized in the cubic phase.
Stabilization of the cubic phase by incorporation of CaO, MgO,
the critical slag layer thickness for the change from passive to
active oxidation was calculated as 155 µm. Based upon the
various assumptions involved in the calculation, this is very
close to the experimentally determined value of 100 µm.
Reaction bonded SiC (RB SiC) is produced by the reaction of
either liquid or gaseous silicon or SiO with carbon in a silicon
carbide/carbon compact. This results in a porous body with
continuous silicon carbide phase; however, these pores can be
filled with nonreacted Si (2–10%) yielding a dense product that
results in excellent mechanical properties. The excellent
wettability between Si and SiC allows this to be carried out for
RB SiC but not for RBSN (silicon nitride). This interpenetrating
grain boundary phase of silicon metal limits the high-temperature
mechanical properties to the melting point of Si (1410°C). When
exposed to aggressive environments, the silicon may be attacked
relatively easily, leading to degradation of properties.
Alumina
Alumina has been shown by Sato et al. [5.33] to dissolve in
aqueous solution at 150–200°C containing NaOH by the
following reactions:
(5.3)
(5.4)
with the second reaction being the faster of the two. Although
no surface interfacial layers were reported, AlOOH solid formed
as part of the overall reaction as shown above. The rate of
dissolution was linearly proportional to the NaOH
concentration. Because the samples of Sato et al. were impure,
containing a silicate grain boundary phase (7% and 0.5%), the
grain boundaries exhibited enhanced corrosion.
The ligand-enhanced dissolution of alumina was discussed
by Brady and House [5.34]. They reported that the dissolution
was dependent upon the structure of the ligand; the oxalate
five-membered ring was the most corrosive, with the malonate
six-membered ring being intermediate, and the succinate seven-
membered ring being the least corrosive. These three ligands
are all adsorbed through two functional groups (i.e., bidentate
adsorption) to a single site. Benzoate, which is adsorbed through
only one functional group (i.e., unidentate adsorption), was
even less corrosive than those adsorbed through two functional
groups.
Silica and Silicates
The dissolution at room temperature and pH=7 of the various
forms of silica has been reported to be a function of the silica
tetrahedra packing density by Wilding et al. [5.35]. Thus the
dissolution increases in the sequence: quartz, cristobalite, opal,
*
The fused silica was tested in a buffered solution of distilled, deionized
water. Different inorganic and organic buffers were used for each pH range.
FIGURE 5.5 Dissolution rates of quartz at 25 and 60°C, and SiO2 glass
at 65°C. (From Ref. 5.34. Copyright CRC Press, Boca Raton, FL.)
positive end vs. the negative end, whereas for Y- and Z-cut
samples both ends exhibited similar rates. Both the X- and Y-
cut ends exhibited dissolution rates much lower than the rate
for Z-cut samples. The rates of dissolution were reported to be
HF concentration dependent and surface reaction controlled.
The substitution of HF molecules in solution with a surface
complex ion was suggested as the surface reaction. The
differences among the various X-, Y-, or Z-cut samples was
attributed to a difference in the number of reactive or kink
sites, rather than a difference in structure for the various
crystallographic faces.
An area of extreme importance that has received little
attention until recently is the relationship between the surface
activity of minerals and their toxicity to humans. Most of the
biological studies of toxicity examine only the effects of particle
size and shape and the mass concentration. An excellent review
of the effects of inhaled minerals was recently given by Guthrie
(5.6)
where:
d = fiber diameter (m)
k = rate constant for silica dissolution (mol/m2s)
Vm= volume of 1 mol of silica in chrysotile (equal to 5.4×
10-5 m3/mol)
A 1-µm-diameter chrysotile fiber will take approximately 9
months to dissolve (based upon the silica concentration of the
fluid at a pH of 4–7 and 37°C). This lifetime is an order-of-
magnitude less than the time required for the onset of diseases’
symptoms. Thus, any biological model must explain this
difference. By comparison, a fused silica fiber of 1-µm diameter
*
Sulfuric acid has been reported to make up 60–70% of the acidity in acid
rain for the northeastern United States.
or
(5.8)
Zirconia-Containing Materials
The hydrothermal effect of water upon the dissolution of yttria
(14 mol%)-stabilized zirconia (YSZ) single crystals was
investigated by Yoshimura et al. [5.46]. They found four regimes
of behavior for YSZ treated at 600°C and 100 MPa for 24 hr,
depending upon the pH of the solution. In alkali solutions (those
containing LiOH, KOH, NaOH, or K 2 CO 3 ), partial
decomposition and dissolution/precipitation were found, with
yttria being the more soluble component. In acidic solutions
(those containing HCl or H2SO4), rapid dissolution of yttria
occurred forming an interface of polycrystalline monoclinic
ZrO2. In reactions with H3PO4 solution, the interface layer
formed was ZrP2O7. In neutral solutions, the dissolution was
minimal.
Superconductors
Murphy et al. [5.47] reported that reaction with water liberated
oxygen, formed Y2BaCuO5 and CuO, in addition to barium
hydroxide, and was a function of temperature and surface area.
This is similar to the leaching of barium from perovskites in
aqueous solutions when the pH is less than 11.5 reported by
Myhra et al. [5.48].
Titanates and Titania
A crystalline titanate mineral assemblage called SYNROC*
has been under investigation for many years as a possible
encapsulant for high-level radioactive wastes. The titanates are
commonly a mixture of perovskites, CaTiO3 and BaTiO3,
zirconolite, CaZrTi2O7, and hollandite, BaAl2Ti6O16. In a study
of the dissolution of these titanates in CO2 enriched (4 ppm)
deionized water (pH=5–6) at 300 and 350°C and 500 bars,
Myhra et al. [5.48] reported the following reactions:
(5.11)
(5.12)
(5.13)
(5.14)
(5.15)
(5.16)
*
SYNROCK was developed in 1978 at the Australian National University
by Prof. T.Ringwood. Several compositional variants have been developed,
called SYNROCK-C, -D, or -F. Additional information can be obtained
from the website www.uic.com.au) of the Uranium Information Centre,
Ltd. in Melbourne, Australia.
The (SO4)- that forms is an active species that reacts with titania
forming etch pits. This phenomenon is called photoelectro-
chemical aging.
and
(5.19)
Oxides
The importance of molten salt reactions is well known in
alumina reduction cells for the production of aluminum metal
(Hall-Heroult process). In this process, the electrolyte consists
of a solution of alumina (<10 wt.%) dissolved into molten
cryolite (Na3AlF6) [5.61]. Pure molten cryolite contains ,
, F-, and Na+ ions. When alumina is added, the complex
ion (x=3–5) forms in addition to the others. In a
study of the cryolite-mullite and cryolite/sodium fluoride-mullite
systems, Siljan and Seltveit [5.62] reported that materials with
high Si/Al ratios experienced high weight losses when in contact
with NaF-cryolite eutectic melts because of the formation of
gaseous SiF4. They reported that mullite readily dissolved into
cryolite and cryolite-NaF melts, and that NaF reacted with
alumina to form β-alumina. Allaire [5.63] studied the resistance
of 15 different commercial refractories with varying Al2O3/SiO2
ratios in a simulated alumina reduction cell and found that the
resistance to attack by molten metallic sodium and NaF increased
with alumina content. He reported that above 700°C, an Al2O3/
SiO2 ratio greater than 0.90 should be used.
The corrosion of fused silica by molten sodium sulfate in
atmospheres containing either 1% SO2/O2 or pure oxygen at
700 and 1000°C has been described by Lawson et al. [5.64] to
take place according to the ease of sodium diffusion in the
various phases that form. Sodium diffuses into the fused silica,
leading to the nucleation of cristobalite. Once a continuous
layer of cristobalite formed, sodium diffusion was minimized.
The sodium at the cristobalite/fused silica interface then diffused
*
Compare with the discussion in the next section involving equations 5.20
through 5.24 and the relationship of soda activity with SO3 levels.
(5.21)
*
The more reduced forms of the sodium-sulfur compounds are the basis of
the amber color formed in the manufacture of brown bottles.
(5.25)
was the faster mechanism. Sato et al. [5.75] reported that this
inner protective layer of silica formed on pressureless sintered
Si3N4 containing 5 wt.% Y2O3 and Al2O3 in contact with molten
potassium sulfate at 1200°C, when the tests were conducted in
air but not when conducted in nitrogen. This same situation
was not true for attack by molten potassium carbonate at 1013°C.
In this case, attack occurred in both nitrogen and air, with air
causing a greater degree of reaction. Compared to studies
performed in molten sodium and lithium sulfate and carbonates,
Sato et al. found that the corrosion rate, in a nitrogen
atmosphere, was independent of the alkali present, with the
sulfates yielding an activation energy of 430 kJ/mol compared
to that of the carbonates of 106 kJ/mol.
Superconductors
In an investigation of the molten-salt synthesis of YBa2Cu3O7-x
(123), Raeder and Knorr [5.78] reported the stability of 123
against decomposition toward several molten salts at 1173 K.
They concluded that 123 was not stable in molten LiCl or the
dichlorides of Cu, Ca, Mg, or Ba, or their mixtures. However,
minimum decomposition was found in the NaCl-KCl system.
The mechanism of decomposition was postulated as being one
of selective dissolution of the barium in the 123 forming BaCl2
and causing the 123 to decompose into several oxide phases
that were consistent with the phase diagram reported by Lee
and Lee [5.79]. These oxide phases were generally CuO and
Y2Cu2O5, or CuO and Y2BaCuO5 depending upon the amount
of barium in the initial mixture.
rich mixed spinel crystals within the magnesia and at the grain
boundaries.
The reaction of silica-containing refractories with molten
iron containing dissolved manganese has been known to be
very deleterious. This reaction, however, is not only a reaction
with a molten metal but also with an oxide of manganese (i.e.,
MnO). The initial reaction between SiO2 and Mn forms MnO
and Si metal. Although this reaction is thermodynamically
unlikely, it has been reported by Kim et al. [5.80] to occur at
1600°C under an argon atmosphere. The subsequent reaction
of MnO and silica can form one of two intermediate compounds,
but more importantly, can form a eutectic liquid with a solidus
temperature of 1250°C.
Many steel plant refractories are carbon-containing from
manufacturing processes involving pitch or tar, organic resins,
or graphite. Solid-solid corrosion through reduction by carbon
takes place, or if the carbon becomes oxidized corrosion by
molten metals is then very seldom a simple reaction involving
only solid ceramic and molten metal.
The attack of molten aluminum upon materials containing
silica follows the reaction:
(5.26)
which should be expected from examination of the free energy
vs. temperature data of an Ellingham diagram. The alumina
that forms in many cases provides an adherent protective layer
against further corrosion [5.81]. This reaction is accompanied
by a volume decrease of about 26% according to Siljan et al.
[5.82]. Although many previous investigators have attributed
the spalling of refractory linings to the volume expansion of
this reaction, Siljan et al. have attributed any expansive spalling
that may occur to the growth of corundum along thermal
gradients and/or any contraction due to the volume decrease.
The action of molten aluminum upon any β-alumina contained
in materials such as high alumina (70%) refractories produces
metallic sodium [5.83]. The metallic sodium present can then
lead to reduction of silica, and if oxidized, it can lead to the
(5.28)
*
Sialon is an acronym used to represent solid solutions of silicon, aluminum,
oxygen, and nitrogen.
5.2.1 Oxides
Alumina
In the ceramics community, alumina is considered one of the
most inert materials toward a large number of environments.
For this reason, alumina that is produced as a 95–100% Al2O3
material is used in many furnace applications. The one area
where its reactivity is often overlooked is its application in
laboratory furnaces. Most high-temperature laboratory furnaces
use alumina as the standard lining. When the materials that
are under investigation react to form gaseous species, and
especially when the furnace atmosphere contains a low partial
pressure of oxygen, one should be aware of the possible reactions
that may occur with alumina.
(5.31)
(5.32)
(5.33)
Alumino-Silicatcs
Arnulf Muan has provided a considerable amount of
experimental data concerning the atmospheric effects upon the
phase equilibria of refractory materials. One such article was
reprinted in the Journal of the American Ceramic Society [5.93]
as a commemorative reprint. This article stressed the importance
of the oxygen partial pressure in determining the phases present
in the reaction of iron oxides with alumino-silicate refractories.
Under oxidizing conditions, large amounts of ferric iron can
substitute for aluminum in the various aluminum-containing
phases; however, under reducing conditions, this substitution is
negligible. Large volume changes accompany some of the phase
changes that occur with damaging results to the refractory. In
addition, the temperature at which liquid phase develops
decreases as the oxygen partial pressure decreases.
Reactions that have occurred between alumino-silicate
refractories and the gaseous exhaust in glass furnace regenerators
(5.36)
The nepheline formed then reacts with SO3 and more Na2O
vapor of the exhaust forming noselite:
(5.37)
Although the precise mechanisms that take place have not been
determined, most likely, the alumina and any silica available
form more nepheline. Large volume expansions (10–15%)
accompany these reactions resulting in spalling or shelling.
Historically, these reactions have been a serious problem to the
glass manufacturer, as they cause plugging of the regenerator
and a less efficient combustion process. Various regenerator
design and material changes have essentially eliminated this
problem; however, it is a reaction that may still occur when the
conditions are appropriate.
Magnesia-Containing Materials
McCauley and coworkers [5.94–5.97] have studied the effects
of vanadium upon the phase equilibria in magnesia-containing
materials. This work was initiated in an effort to understand
the effects of vanadium impurities in fuel oils on basic
refractories. It has been found that only small amounts of V2O5
are needed to alter the phase assemblages in high magnesia
materials. The reactions that occur generally form low melting
vanadates (i.e., tricalcium and trimagnesium vanadates with
melting points of 1380 and 1145°C, respectively, and
magnesium-calcium-vanadium garnet with its melting point of
1167°C) and, depending upon the exact compositions, can
develop appreciable amounts of liquid at service temperatures.
Zirconia
Lepistö et al. [5.99] studied the effects of humid conditions at
150°C for up to 1000 hrs upon the stability of metastable tetragonal
phase in tetragonal zirconia polycrystals (TZP). Several different
materials containing a small amount of yttria were tested, and
all were found to contain increasing amounts of monoclinic
zirconia as the exposure time increased. It was believed that
surface finish along with the grain size and yttria content all had
an effect upon the transformation of the tetragonal-to-monoclinic
zirconia. In a later study, Lepistö and Mäntylä [5.100] concluded
that the stability of yttria-containing TZP in humid atmospheres
was through the dissolution of ZrO2 at grain boundaries with the
subsequent relief of localized stresses, followed by the
transformation of the tetragonal-to-monoclininc phase. The
proposed mechanism of dis- solution was through the dissociation
of adsorbed water molecules on the zirconia surface. The oxygen
ions formed were then proposed to anneal the oxygen vacancies
present within the yttria-containing zirconia. As a result, water
is formed, which then allowed continued dissolution. The water
adsorption was greatest at defect sites, and thus was dependent
upon the yttria concentration. Lepistö and Mäntylä found that
both yttria and alumina increased the dissolution rate, whereas
ceria did not, as it did not change the vacancy concentration.
Thus the tetragonal phase stability toward humid atmospheres
can be increased by using ceria as the dopant.
Silicon Nitride
Oxidation. The oxidation of Si3N4 is dependent upon the
manufacturing process used: chemically vapor-deposited (CVD)
materials exhibit the slowest and smallest amount of oxidation
(5.41)
Other Nitrides
The crystalline solution series of materials of alumina dissolved
into β-silicon nitride (Si6-xAlxOxN8-x) make up a truly interesting
series of materials. The hopes were that these materials would
yield properties that were the best of the two end members.
One improvement over silicon nitride is the oxidation resistance
with increasing amounts of alumina while maintaining the
relatively low thermal expansion characteristics of the pure
silicon nitride phase. Weight gain behaviors have been reported
to be parabolic with mullite being the oxide that formed on the
surface. Singhal and Lange [5.125] reported that mullite formed
only in those compositions containing more than 20 wt.%
alumina and that, above 40 wt.% alumina, additional
unidentified phases occurred. Chartier et al. [5.126] prepared
sialon crystalline solutions with x=0.4. Because 14.05 wt.%
yttria was added to the original mix, the final pressureless
sintered samples contained β’-Y2Si2O7 and a glassy phase as
grain boundary phases. Oxidation was parabolic and very slow
below 1380°C. Above this temperature, more rapid oxidation
occurred with departures from parabolic behavior. A thin
alumino-silicate film formed first, but as metal cation migration
occurred (predominately yttrium) reaction with this film formed
more complex silicates. This film was dense below 1400°C,
and gradually became porous and nonprotective as the
temperature was increased. Visual observation indicated a light
gray zone under the surface scale that Chartier et al. reported
to be attributable to selective oxidation of the grain boundary
phase.
Wang et al. [5.127] investigated the oxidation in air of rare-
earth aluminum oxynitrides with the ideal formula LnAl12O18N
(Ln=La, Ce, Pr, Nd, Sm, and Gd) at temperatures ranging from
700 to 1500°C. Noticeable oxidation started at about 700°C,
and increased with temperature. The final reaction products
depended upon the particular rare-earth, but progressed through
several intermediate stages. At temperatures around 1000°C,
the products were LnAl11O18 and α-alumina. LnAlO3 also formed
with or without the disappearance of LnAl11O18 at higher
temperatures depending upon the rare-earth. In cerium-
containing materials, CeO2 formed at 900°C. At temperatures
as low as 700°C, lattice parameter changes (increasing c/a ratio)
were noted for the oxynitrides, which were attributed to the
initiation of oxidation.
AlN is an important material in the electronic ceramics
industry and is an example of when a small amount of oxidation
is beneficial to the application. In this case, the formation of a
thin (1–2 µm) protective coating of AlON is formed and is used
to improve the adhesion of copper films. Suryanarayana [5.128]
found the oxidation of AlN powders between 600 and 1000°C
in flowing air to follow a linear rate law initially, and then a
parabolic law as the oxide layer thickness became sufficient to
require diffusion for further growth. In contrast, Abid et al.
[5.129] found that the oxide layer that formed on polycrystalline
AlN in air at 1200°C was α-Al2O3, whereas below 800°C, no
oxidation was observed. Dutta et al. [5.130] reported that
oxidation of sintered polycrystalline AlN between 20 and 200°C
progressed from individual α-Al2O3 particles of 2–3 nm in size
to a 50-nm-thick film after 150 hr at 200°C. They also
commented that their data were consistent with the formation
of an oxynitride layer, but believed α-Al2O3 to be the oxide
formed at low temperature based on thermodynamic calculations.
Others have shown that an oxynitride formed as an intermediate
preceding alumina formation at high temperatures [5.131], and
Silicon Carbide
Oxidation. The oxidation of green hexagonal powdered SiC
has been described by Ervin [5.134]. Ervin stated that oxidation
at low oxygen pressures took place with the formation of SiO
gas, while at atmospheric pressure under flowing air, SiO2
formed. The rate-controlling step was thought to be the growth
of an ordered lattice of SiO2 by solid diffusion. The following
reactions are representative of the oxidation of silicon carbide:
(5.45)
(5.46)
the thick oxide on the carbon face initially grew with linear
kinetics but then changed to parabolic when the thickness
became greater than 250 nm. At high temperatures and/or
long times during oxidation of powdered samples, the
oxidation rate changed from parabolic to linear presumably
because the growth of the linearly controlled face overtook
that of the parabolically controlled face. This change in
oxidation rate at high temperatures has been attributed to a
change in the oxide layer from amorphous to crystalline by
Ervin [5.134] and Jorgensen et al. [5.137], and suggested by
Costello and Tressler [5.138].
The desorption of CO gas formed at the SiC/SiO2 interface
has been reported to be the rate-controlling step by Singhal
[5.139]; however, Hinze et al. [5.140] and many others have
reported that it is the inward diffusion of oxygen through the
surface layer of SiO2. Spear et al. [5.141] ruled out the diffusion
of CO as rate-controlling based upon their experiments that
exhibited a dependence of the oxidation rate upon the partial
pressure of oxygen and the almost identical activation energies
obtained for the oxidation of SiC and Si metal. Fergus and
Worrell [5.142] have concluded that the various contradictions
in reported kinetics were attributable to a change in the diffusing
species from molecular to ionic oxygen at about 1400°C. This
was based upon two observations: one being that the activation
energy for the growth of amorphous silica on CVD SiC increased
above 1400°C, and the other being that the activation energy
for the growth of cristobalite increased, but at the higher
temperature of 1600°C. Decreases in oxidation rates at low
temperatures have been attributed to sufficiently long times to
allow crystallization of the silica scale.
In an analysis of the various possible rate-controlling steps,
Luthra [5.85] concluded that a mixed interface reaction/diffusion
process was the limiting feature in the oxidation of SiC. This
was based upon the following facts:
1. Oxidation rate is lower than for pure silicon
2. Presence of gas bubbles in the oxide layer
(5.48)
Other Carbides
Probably the next most important carbide after silicon carbide
is tungsten carbide. The principal application of this material
is in cemented carbide cutting tools. The carbides of titanium,
tantalum, and niobium are used as alloying additions to WC.
Addition of TiC to WC cutting tools causes the formation of a
titanium oxide surface layer that greatly increases the tool’s
wear resistance. When WC oxidizes, it forms the volatile WO3
oxide that offers no protection to wear.
Most of the oxidation studies conducted on TiC are rather
dated; however, reasonable agreement exists among the various
studies. Stewart and Cutler [5.156] found that the oxide layer
that formed below 400°C was anatase and that above 600°C it
was rutile. Single crystal studies indicated no difference in
oxidation between the (100) and (110) faces at 1000°C. At low
temperatures (752–800°C), the rate of oxidation exhibited a
dependence upon the oxygen partial pressure to the 1/6 power,
whereas at high temperatures, the dependence was to the 1/4
power. The actual mechanism of oxidation appeared to be
mixed with a near-parabolic rate initially changing to a near-
linear rate at longer times.
The use of ZrC at high temperatures (>450°C) has been limited
due to excessive oxidation, although it possesses other excellent
properties. The oxidation of powdered ZrC at low temperatures
5.2.3 Borides
Several of the diborides are of considerable interest because of
their high melting points and high strengths at elevated
temperatures. Probably the one that has received the most
attention is TiB2; however, ZrB2, HfB2, NbB2, and TaB2 are
also of interest. These are the most attractive because of their
high stability compared to the other diborides. Like the carbides
(5.50)
(5.51)
5.2.4 Silicides
The oxidation of MoSi2 has been reported to occur by several
mechanisms by Fitzer [5.163] depending upon the temperature
and the oxygen partial pressure. Initially, only MoO3 formed,
but volatilized, allowing the formation of SiO2. The partial
pressure of oxygen at the interface MoSi2/SiO2 then decreased,
allowing the oxidation of only silicon to continue. At very high
temperatures (>1200°C) and low oxygen pressures (<10-6 atm),
active oxidation occurred with the formation of volatile silicon
monoxide, as long as the silicon content on the surface was
sufficient. At low pressures of oxygen, selective oxidation of
silicon occurred because of its greater affinity for oxygen than
molybdenum. The selective oxidation of silicon led to the
formation of a sublayer of Mo 5Si 3. At moderately high
temperatures (around 1000°C) and high oxygen pressures (≈10-
2
atm), the evaporation of the molybdenum oxides formed led
to a protective SiO2 layer. During the volatilization of the
molybdenum oxides, the SiO2 layer was very porous allowing
rapid oxidation with temperature increase. At lower
temperatures, where the molybdenum oxides did not volatilize
but remained as solid oxide reaction products, a continuous
silica layer could not form. This occurred at temperatures below
600°C, and is called pesting, which can lead to total destruction
of the material. The actual amount of pesting that occurred,
however, was dependent upon the microstructure [5.164]. Lin
et al. [5.165] found that the oxidation of MoSi2 in a combustion
gas environment (i.e., natural gas) was very similar to that
described above. Initially, SiO2 and volatile MoO3 formed until
the silica layer was of sufficient thickness to diminish the H2O/
CO2 activity (equivalent to a low partial pressure of oxygen) to
a level where Mo5Si3 formed. The change in mechanisms took
only minutes at temperatures of 1370 and 1600°C.
5.2.5 Superconductors
It is well known that YBa2Cu3Ox (123) is unstable with respect to
reaction with carbon dioxide or water vapor. This reaction is related
to the large enthalpies of formation of barium carbonate and barium
hydroxide, which are -64.4 and -35.4 kcal, respectively. Davison
et al. [5.166] reported the formation of barium carbonate when
123 was held over water for 48 hr. Yan et al. [5.167] reported the
formation of Cu(OH)2 when 123 was exposed to 85°C and 85%
relative humidity for 90 min. After exposing 123 to 80°C and
100% relative humidity for times ranging from 15 min to 24 hr,
Fitch and Burdick [5.168] reported the formation of Y2BaCuOx,
BaCO3, CuO, Cu(OH)2, and the possible formation of Y(OH)3
and BaO. Fitch and Burdick, who noticed that corrosion was visibly
present by a significant expansion of their samples, concluded that
barium was leached first then reacted with atmospheric CO2 to
form BaCO3 on the surface.
5.3.1 Silica
Miller et al. [5.170] have shown that carbon reacts with SiO2
to form the intermediate phase SiC, which then reacts with
silica to form the gaseous phase SiO. The following equations
were given to represent the reaction:
(5.52)
(5.53)
5.3.2 Magnesia
Magnesia vaporization is important in basic refractories where
it migrates to form a region rich in magnesia by vaporization
and condensation, and leaving behind a region of high porosity.
The zone of high porosity causes a mechanically weak area
that may crack or spall. Vaporization and condensation of
magnesia can also occur in silicon nitride where it is used as a
sintering aid (see discussion on page 203).
In pitch-containing high magnesia refractories, it has been
found [5.171] that the carbon in the refractory can react with
the magnesia to form magnesium gas according to the following
equation:
(5.54)
5.3.3 Superconductors
The YBa2Cu3Ox (123) superconductors, where x=6.5–7.0, have
been reported to exhibit reaction and/or decomposition when
or
(5.57)
they reported that the nine elements Ru, Rh, Pd, Ag, Os, Ir, Pt,
Au, and Hg should not react with CuO, and most likely would
not react with 123. Murphy et al. [5.47] reported that 123 was
nonreactive toward silver and to a lesser extent, gold.
One of the potential applications of superconductors is that
of thin films on a semiconductor substrate; however, the most
widely used semiconductor substrate material, silicon, reacts
with 123. An examination of the various phase equilibria
indicated that BaSi2O5 does not react with 123, because these
two materials form a stable tie-line in the BaO–Y2O3–CuO–
SiO2 quaternary system. Thus this barium silicate could be used
as a buffer layer for production purposes or during manufacture
of thin films on semiconducting substrates.
Mikalsen et al. [5.173] reported that no reaction occurred
between thin film superconductors in the Bi–Sr–Ca–Cu–O system
and MgO substrates even after annealing at 850°C for 30 min.
Thin films on Al2O3, however, reacted and became insulating
and transparent. Abe et al. [5.174] and Ibara et al. [5.175] have
reported that melts of BiSrCaCu2Ox or BiPbySr CaCu2Ox reacted
with alumina crucibles contaminating their samples.
REFERENCES
5.1. Sandhage, K.H.; Yurek, G.J. Indirect dissolution of (Al,
Cr)2O3 in CaO–MgO–Al2O3–SiO2 (CMAS) melts. J. Am.
Ceram. Soc. 1991, 74 (8), 1941–1954.
5.2. Oishi, Y.; Cooper, A.R., Jr.; Kingery, W.D. Dissolution in
6.1 INTRODUCTION
The corrosion of glassy materials is predominantly through
the action of aqueous media. The attack by gases quite often is
that of water vapor or some solution after various species
condense and dissolve in the water. Therefore this chapter will
be devoted mostly to aqueous attack.
In general, very high silica (>96% SiO2), aluminosilicate,
and borosilicate compositions have excellent corrosion
resistance to a variety of environments. Silicate glasses, in
257
Copyright © 2004 by Marcel Dekker, Inc.
258 Chapter 6
Chapter 6
Copyright © 2004 by Marcel Dekker, Inc.
Corrosion of Specific Glassy Materials
A weight loss of 1 mg/cm2 is equivalent to a depth loss of 0.01 mm/(specific gravity of glass) for those cases where the attack is not
selective. (From Ref 6.1, Copyright © 1966 by John Wiley & Sons, Inc. This material is used by permission of John Wiley & Sons,
Inc.)
261
Copyright © 2004 by Marcel Dekker, Inc.
262 Chapter 6
into the glass and the constituents of the glass diffuse into the
tin. The reaction zone in the glass is about 25 µm thick. Many
investigators have studied the tin oxide gradient of float glass
and have reported a rather complex behavior [6.45–6.54].
Stannous tin is dominant at the near surface. A typical hump
occurs in the tin profile at between 5 and 10 µm where stannic
(or oxidized tin) is predominant. This hump has been attributed
to the additional tin from the ion exchange with calcium by
Franz [6.55]. Investigation of extremely thin layers of glass
has indicated tin oxide contents as high as 36% at the surface
[6.52] (see Fig. 6.2). The amount of the tin contained within
the glass surface and the depth to which it penetrates is
dependent upon the exposure time and temperature (which
relates to glass production tonnage and thickness), and the
amount and type of impurities (especially sulfur) contained in
the tin. Thin glass travels through the bath faster than thick
glass and therefore has less time for the various reactions to
take place.
At the hot end of the bath, iron oxide in the glass will migrate
toward the bottom surface where it is reduced (by reaction
with either stannous oxide or hydrogen) to iron metal and
dissolves into the tin. At the cooler end of the bath, this tin/
iron alloy will oxidize (oxygen coming from air ingress) forming
both iron oxide and tin oxide. Iron has a greater potential to
oxidize than tin and therefore acts as a scavenger for oxygen.
* The deposit of calcium oxide reacts with atmospheric carbon dioxide forming
calcium carbonate on the glass surface that is insoluble in water and must be
washed off with a vinegar solution.
(6.7)
or:
(6.8)
As pH increased from 10 to 14, the rate of attack increased
about 400 times. Higher iodine contents lowered the durability.
REFERENCES
6.1. Hutchins, J.R. III; Harrington, R.V. Glass. In Encyclopedia
of Chemical Technology; 2nd Ed; Wiley: New York, 1966;
Vol. 10. 572.
6.2. Wang, F.F-Y.; Tooley, F.V. Influence of reaction products on
reaction between water and soda-lime-silica glass. J. Am.
Ceram. Soc. 1958, 41(12), 521–524.
6.3. Brady, P.V.; House, W.A. Surface-controlled dissolution and
growth of minerals. In Physics and Chemistry of Mineral
Surfaces, Brady, P.V., Ed.; CRC Press: New York, 1996; 225–
305. Chap. 4.
6.4. Simpson, H.E. Study of surface structure of glass as related
to its durability. J. Am. Ceram. Soc. 1958, 41 (2), 43–49.
6.5. Chen, H.; Park, J.W. Atmospheric reaction at the surface of
sodium disilicate glass. Phys. Chem. Glasses 1981, 22 (2),
39–42.
6.6. Charles, R.J. Static fatigue of glass: I.J. Appl. Phys. 1958, 29
(11), 1549–1553.
6.7. Scholze, V.H.; Helmreich, D.; Bakardjiev. Investigation of the
behavior of soda-lime-silica glasses in dilute acids (Gr). Glass
Technol. Ber. 1975, 48 (12), 237–247.
6.8. Sanders, D.M.; Hench, L.L. Mechanisms of glass corrosion.
J. Am. Ceram. Soc. 1973, 56 (7), 373–377.
6.9. Schmidt, Yu.A. Structure of Glass; Consultants Bureau: New
York, 1958; Vol. 1. Translated from the Russian.
6.10. Boksay, Z.; Bouquet, G.; Dobos, S. Diffusion processes in
the surface layer of glass. Phys. Chem. Glasses 1967, 8 (4),
140–144.
6.11. Doremus, R.H. Interdiffusion of hydrogen and alkali ions in
a glass surface. J. Non-Cryst. Solids 1975, 19, 137–144.
6.12. Boksay, Z.; Bouquet, G.; Dobos, S. The kinetics of the
formation of leached layers on glass surfaces. Phys. Chem.
Glasses 1968, 9 (2), 69–71.
6.13. Das, C.R. Reaction of dehydrated surface of partially leached
glass with water. J. Am. Ceram. Soc. 1979, 62 (7–8), 398–402.
6.14. Rana, M.A.; Douglas, R.W. The reaction between glass and
7.1 INTRODUCTION
Although the term composite historically meant any product
made from a combination of two (or more) materials, the
modern meaning is less broad in scope. In general, a composite
is manufactured in an attempt to obtain the best properties of
two materials or at least to capture a specific property of each
material that is potentially better in the composite. It is also
possible for the composite to have a particular property that
neither component exhibited individually. According to Holmes
and Just [7.1], a true composite is where distinct materials are
combined in a nonrandom manner to produce overall structural
characteristics superior to those of the individual components.
Although, in a very broad sense, products such as glazed
291
Copyright © 2004 by Marcel Dekker, Inc.
292 Chapter 7
will not be able to propagate through it. The matrix must not
damage the reinforcement and it must transmit any stresses to
the reinforcement. Thus the adhesion of the matrix to the
reinforcement is of prime importance for mechanical integrity
and is the region of greatest importance related to corrosion.
Since it is necessary to have weak interfaces to maximize
toughness (i.e., resistance to crack propagation), the
development of optimum fiber/matrix interfaces is quite
difficult. To obtain these optimum characteristics, it is
sometimes required to coat the reinforcement fibers with
various materials to obtain the proper debonding, sliding, and/
or reaction characteristics. Fibers that do not debond do not
enhance toughening and lead only to increased brittle fracture
of the composite [7.2–7.7].
A recent development in composites is that of a nanosized
second phase or reinforcement material. The second phase
particles are generally less than 300 nm and are present in
amounts equal to 1–30 vol.%. These new composites
unfortunately have been called nanocomposites.
Before going into the specifics of corrosion of composite
materials, a few words must be said about those materials that
have been called cermets. Historically, the term cermet was
derived to cover those materials composed of cobalt-bonded
tungsten carbide and used as cutting tools. Since cermets
contain both ceramics and metals, some confusion has existed
in the literature as to an exact meaning. The term, however,
has been used to cover a broad list of materials. It appears that
the ceramic community confines cermets to essentially cutting
tool materials, whatever the matrix or reinforcement, whereas
the metals community confines cermets to only those materials
with a metal matrix. Since the broader concept of composites
includes those materials called cermets, only the term composite
will be used in the discussion below.
The actual corrosion of composite materials quite often
begins with reaction of the reinforcement material and
especially with any interface material (called the interphase)
used to coat the reinforcement for debonding. One property
7.2 REINFORCEMENT
7.2.1 Fibers
Various types of materials have been used as the fibrous
reinforcement. These include various glasses, metals, oxides,
(7.1)
7.2.3 Particulates
For a discussion of the various mechanisms involved in
toughening composites when particulates are used as the
* Cooper and Hall use the term fluorophlogopite interchangeably with the
terms mica, fluoromica, and fluorophyllosilicate, which may cause some
confusion unless the reader is well versed in mineralogy.
Al2O3-Matrix Composites
A BN/SiC-coated Nicalon™ fiber/Al2O3 composite has been
reported by Heredia et al. [7.29] to become embrittled when heated
to temperatures between 650°C and 850°C. The tensile strength
after heat treatment for 24 hr was reduced by one-half. Only the
outer regions of the ~3-mm-thick samples were fully embrittled.
A 21 vol.% SiC in alumina composite was reported by
Borom et al. [7.30] to form a reaction zone upon oxidation at
1530°C for 150 hr that contained mullite and an amorphous
aluminosilicate phase containing bubbles from the formation
of CO. The SiO2 formed by the oxidation of the SiC reacted
with the alumina matrix to form the mullite. It is important
that the formation of silica in the outer layer is sufficient for
complete conversion of the alumina to mullite [7.31].
Insufficient silica causes a rigid scale that delaminates. Too
much silica forms a scale containing mullite and silica on an
alumina substrate that may also delaminate due to expansion
mismatch during thermal cycling. A matrix of mullite works
much better than alumina since the scale is more compatible
with the substrate, both containing mullite, and thus forms a
protective layer. Luthra [7.32] reported that the products of
reaction of SiC with alumina should be mullite and alumina
when the SiC content is below 24.4 vol.% and silica and mullite
when it is greater than 24.4 vol.%. In practice, this limit will
vary due to mullite forming over a range of compositions.
A TiN/Al2O3 composite was reported by Mukerji and Biswas
[7.33] to exhibit linear oxidation kinetics above 820°C after a
* Both these phases contain the same elements but in different ratios (sapphirine
M4A5S2 and cordierite M2A2S5) and in small quantities could easily be confused
with one another by EDS analysis.
(7.10)
The latest preference for SiC/SiC composite is one with fibers
of improved microstructure and chemistry called Sylramic™*
*
†
Sylramic™, Dow Corning Corp., Midland, MI.
Pesting was originally used to describe the formation of a powder-like
deposit on the surface of metallic silicides during oxidation; however, it is
now used to describe a similar phenomenon on any material.
* The Pilling and Bedworth coefficient is defined as the ratio of the volume of
silica formed to the volume of the ceramic consumed. See N.G. Pilling and
R.E. Bedworth, J. Int. Met., 29, 529 (1923).
REFERENCES
7.29. Heredia, F.E.; McNulty, J.C.; Zok, F.W.; Evans, A.G. Oxidation
embrittlement probe for ceramic matrix composites. J. Am.
Ceram. Soc. 1995, 78 (8), 2097–2100.
7.30. Borom, M.P.; Bolon, R.B.; Brun, M.K. Oxidation mechanism
of MoSi2 particles in mullite. Adv. Ceram. Mater. 1988, 3 (6),
607–611.
7.31. Borom, M.P.; Brun, M.K.; Szala, L.E. Kinetics of oxidation of
carbide and silicide dispersed phases in oxide matrices. Adv.
Ceram. Mater. 1988, 3 (5), 491–497.
7.32. Luthra, K.L. Oxidation of SiC-containing composites. Ceram.
Eng. Sci. Proc. 1987, 8 (7–8), 649–653.
7.33. Mukerji, J.; Biswas, S.K. Synthesis, properties, and oxidation
of alumina-titanium nitride composites. J. Am. Ceram. Soc.
1990, 73 (1), 142–145.
7.34. Tampieri, A.; Bellosi, A. Oxidation resistance of alumina-
titanium nitride and alumina-titanium carbide composites.
J. Am. Ceram. Soc. 1992, 75 (6), 1688–1690.
7.35. Revankar, V.; Hexemer, R.; Mroz, C; Bothwell, D.; Goel, A.;
Bray, D.; Blakely, K. Novel process for titanium nitride
whisker synthesis and their use in alumina composites. In
Advanced Synthesis and Processing of Composites and
Advanced Ceramics, Ceramic Transactions; Logan, K.V., Ed.;
American Ceramic Society: Westerville, OH, 1995; Vol. 56,
135–146.
7.36. Wang, T.C.; Chen, R.Z.; Tuan, W.H. Oxidation resistance
of Nitoughened Al2O3. J. Eur. Ceram. Soc. 2003, 23, 927–
934.
7.37. Nelson, H.G. A challenge to materials: Advanced hypersonic
flight hydrogen and high temperature materials. In Proc. 1993
Conf. on Processing, Fabrication and Applications of
Advanced Composites, Long Beach, CA; Upadhya, K., Ed.;
ASM: Ohio, Aug 9–11, 1993; 11–20.
7.38. Sambell, R.A.; Bowen, D.; Phillips, D.C. Carbon fiber
composites with ceramic and glass matrices, Part 1,
Discontinuous fibers. J. Mater. Sci. 1972, 7, 663–675.
7.39. Brennan, J.J. Interfacial characterization of glass and glass-
ceramic matrix/Nicalon SiC composites. In Tailoring
8.1 INTRODUCTION
Probably the most important property that is affected by
corrosion is that of mechanical strength. Other properties are
also affected by corrosion; however, they generally do not lead
to failure, as is often the case with changes in strength. Strength
loss is not the only mechanical effect of corrosion, because
there are also many cases where the effects of corrosion lead
to increased strength. Increases in strength due to corrosion
are the result of healing of cracks and flaws in the surface
layers of a specimen due, quite often, to the diffusion of im-
purities from the bulk to the surface. This change in chemistry
333
Copyright © 2004 by Marcel Dekker, Inc.
334 Chapter 8
8.2 MECHANISMS
FIGURE 8.1 Crack velocity vs. applied force (or KI). (From Ref. 8.13.)
force, with the exact position of the curve and its slope being
dependent on humidity. At higher humidities, the crack growth
occurs more rapidly and at a lower force. In addition,
Wiederhorn and Bolz [8.14] have shown that the slope and
position of the curve in this first region is dependent upon the
glass composition (stress corrosion resistance was in the
order fused silica>aluminosilicate>borosilicate>soda-lime
silicate>lead silicate), and Wiederhorn and Johnson [8.15] have
shown that it is dependent upon the pH. In the second region,
the crack velocity is essentially independent of the applied force.
At higher humidities, this portion of the curve shifts to higher
velocities. In the third region, the crack velocity is again
dependent upon the applied force, but the slope is much steeper,
indicative of a different mechanism for crack propagation. This
third region is also independent of the humidity.
(8.2)
where:
P=applied load
L=crack length
w=total thickness
a=web thickness
t=half-width
The actual shape of the velocity vs. KI curves is determined by
a balance between the corrosion process, which tends to
increase the crack tip radius, and the stress-corrosion process,
which tends to decrease the crack tip radius [8.17].
Wiederhorn et al. [8.18] gave an equation of the following
type for determining the crack velocity in aqueous media:
(8.3)
where:
v=crack velocity
vO=empirical constant
aH2O=activity of water
G*=free energy of activation
R=gas constant
T=temperature
derived from reaction rate theory, assuming that crack velocity
was directly proportional to the reaction rate. In addition, they
where:
V=crack velocity
Vo=empirical constant
KI=applied stress intensity
n=slope of the exponential plot
FIGURE 8.3 Fracture strength vs. time for Si3N4 and SiC. (From Ref.
8.26.)
where:
Sf=fracture strength
A’=material constant
s=loading rate
n=stress corrosion susceptibility constant
* The application of the Student’s t-test can be found in any elementary statistics
book.
Bioactive Materials
Bioactive ceramics include those materials that rapidly react
with human tissue to form direct chemical bonds across the
interface. Poor bonding across this interface and a sensitivity
to stress corrosion cracking has limited the use of some
FIGURE 8.5 Applied stress vs. time-to-failure; the so-called fatigue knee.
REFERENCES
8.1. Jakus, K.; Ritter, J.E. Jr; Sullivan, J.M. Dependency of fatigue
predictions on the form of the crack velocity equation. J.
Am. Ceram. Soc. 1981, 64 (6), 372–374.
8.2. Matthewson, M.J. Models for fiber reliability. Proc. Int. Symp.
Fiber Optic Networks & Video Communications: Berlin,
Germany, Apr., 1993.
8.3. Johnson, S.M.; Dalgleish, B.J.; Evans, A.G. High temperature
failure of polycrystalline alumina: III. Failure times. J. Am.
Ceram. Soc. 1984, 67 (11), 759–763.
8.4. Evans, A.G.; Blumenthal, W. High temperature failure
mechanisms in ceramic polycrystals. In Deformation of
Ceramics II; Tressler, R.E., Bradt, R.C. Eds.; (1984). Plenum
Publishing Co.: New York, 1984; 487–505.
8.5. Cao, H.C.; Dalgleish, B.J.; Hsueh, C-H.; Evans, A.G. High-
temperature stress corrosion cracking in ceramics. J. Am.
Ceram. Soc. 1987, 70 (4), 257–264.
8.6. Lange, F.F. High-temperature strength behavior of hot-pressed
Si3N4: evidence for subcritical crack growth. J. Am. Ceram.
Soc. 1974, 57 (2), 84–87.
8.7. Lange, F.F. Evidence for cavitation crack growth in Si3N4. J.
Am. Ceram. Soc. 1979, 62 (3–4), 222–223.
8.8. Chuang, T-J. A diffusive crack-growth model for creep
fracture. J. Am. Ceram. Soc. 1982, 65 (2), 93–103.
8.9. Hillig, W.B.; Charles, R.J. Surfaces, stress-dependent surface
reactions, and strength. In High Strength Materials; Zackey,
V.F., Ed.; Wiley Sons: New York, 1965; 682–705. Chapter
17.
8.10. Bando, Y.; Ito, S.; Tomozawa, M. Direct observation of crack
tip geometry of SiO 2 glass by high-resolution electron
microscopy. J. Am. Ceram. Soc. 1984, 67 (3), C36–C37.
8.11. Ito, S.; Tomozawa, M. Crack blunting of high-silica glass. J.
Am. Ceram. Soc. 1982, 65 (8), 368–371.
8.12. Charles, R.J. J. Appl. Phys. 1958, 29, 1549, 1554.
8.13. Wiederhorn, S.M. Influence of water vapor on crack
propagation in soda-lime glass. J. Am. Ceram. Soc. 1967, 50
(8), 407–414.
8.14. Wiederhorn, S.M.; Bolz, L.H. Stress corrosion and static
fatigue of glass. J. Am. Ceram. Soc. 1970, 53 (10), 543–548.
8.15. Wiederhorn, S.M.; Johnson, H. Effect of electrolyte pH on
crack propagation in glass. J. Am. Ceram. Soc. 1973, 56 (4),
192–197.
8.16. Wiederhorn, S.M. A chemical interpretation of static fatigue.
J. Am. Ceram. Soc. 1972, 55 (2), 81–85.
8.17. Simmons, C.J.; Freiman, S.W. Effect of corrosion processes
on subcritical crack growth in glass. J. Am. Ceram. Soc. 1981,
64 (11), 683–686.
8.18. Wiederhorn, S.M.; Freiman, S.W.; Fuller, E.R. Jr.; Simmons,
C.J. Effect of water and other dielectrics on crack growth. J.
Mater. Sci. 1982, 17 (12), 3460–3478.
8.31. Wang, L.; He, C.; Wu, J.G. Oxidation of sintered silicon nitride
materials. In Proceedings of the 3rd International Symposium
on Ceramic Materials and Components for Engines; Tennery,
V.J., Ed.; Am. Ceram. Soc.: Westerville, OH, 1989; 604–611.
8.32. Lange, F.F.; Davis, B.I. Development of surface stresses during
the oxidation of several Si3N4/CeO2 materials. J. Am. Ceram.
Soc. 1979, 62 (11–12), 629–630.
8.33. O’Brien, M.H.; Huang, C.M.; Coon, D.N. Oxidation and
retained strength of in-situ O’–β’ SiAlON composites. Ceram.
Eng. Sci. Proc. 1993, 14 (7–8), 350–357. Part 1.
8.34. Clausen, N.; Wagner, R.; Gauckler, L.J.; Petzow, G. Nitride-
stabilized cubic ZrO2. Presented at the 79th Annual Mtg.
Am. Ceram. Soc., Chicago, Apr 23, 1977; Am. Ceram. Soc.
Bull [abstr]. 1977; 56 (3), 301. Paper no. 136-B-77.
8.35. Lange, F.F. Compressive surface stresses developed in ceramics
by an oxidation-induced phase change. J. Am. Ceram. Soc.
1980, 63 (1–2), 38–40.
8.36. Kawakubo, T.; Komeya, K. Static and cyclic fatigue behavior
of a sintered silicon nitride at room temperature. J. Am.
Ceram. Soc. 1987, 70 (6), 400–405.
8.37. Fett, T.; Himsolt, G.; Munz, D. Cyclic fatigue of hot-pressed Si3N4
at high temperatures. Adv. Ceram. Mater. 1986, 1 (2), 179–184.
8.38. Tajima, Y.; Urashima, K.; Watanabe, M.; Matsuo, Y. Static,
cyclic and dynamic fatigue behavior of silicon nitride. In
Proceedings of the 3rd International Symposium on Ceramic
Materials and Components for Engines; Tennery, V.J., Ed.;
Am. Ceram. Soc.: Westerville, OH, 1989; 719–728.
8.39. Clark, T.J. Fracture properties of thermally aged ceramic fiber
produced by polymer pyrolysis. In Advances in Ceramics,
Fractography of Glasses and Ceramics; Varner, J.R., Frechette,
V.D., Eds.; Am. Ceram. Soc.: Westerville, OH, 1988; Vol.
22, 279–293.
8.40. Butt, D.P.; Tressler, R.E.; Spear, K.E. Silicon carbide materials
in metallurgical heat-treatment environments. Am. Ceram.
Soc. Bull. 1992, 71 (11), 1683–1690.
8.41. Li, C-T.; Langley, N.R. Development of a fractographic
method for the study of high-temperature failure of ceramic
8.52. Frakes, J.T.; Brown, S.D.; Kenner, G.H. Delayed failure and
aging of porous alumina in water and physiological media.
Am. Ceram. Soc. Bull. 1974, 53 (2), 183–187.
8.53. Seidelmann, U.; Richter, H.; Sotesz, U. Failure of ceramic hip
endoprostheses by slow crack growth-lifetime prediction. J.
Biomed. Mater. Res. 1982, 16, 705–713.
8.54. Ritter, J.E. Jr; Greenspan, D.C.; Palmer, R.A.; Hench, L.L.
Use of fracture mechanics theory in lifetime predictions for
alumina and bioglass-coated alumina. J.Biomed. Mater. Res.
1979, 13, 251–263.
8.55. Barry, C.; Nicholson, P.S. Stress corrosion cracking of a
bioactive glass. Ad. Ceram. Mater. 1988, 3 (2), 127–130.
8.56. Troczynski, T.B.; Nicholson, P.S. Stress corrosion cracking
of bioactive glass composites. J. Am. Ceram. Soc. 1990, 73
(1), 164–166.
8.57. Sato, T.; Tokunaga, Y.; Endo, T.; Shimada, M.; Komeya, K.;
Komatsu, M.; Kameda, T. Corrosion of silicon nitride
ceramics in aqueous hydrogen chloride solutions. J. Am.
Ceram. Soc. 1988, 71 (12), 1074–1079.
8.58. Dabbs, T.P.; Lawn, B.R. Strength and fatigue properties of
optical glass fibers containing microindentation flaws. J. Am.
Ceram. Soc. 1985, 68 (11), 563–569.
8.59. Lawn, B.R.; Evans, A.G. A model for crack initiation in elastic/
plastic indentation fields. J. Mater. Sci. 1977, 12 (11), 2195–
2199.
8.60. Lawn, B.R.; Dabbs, T.P.; Fairbanks, C.J. Kinetics of shear-
activated indentation crack initiation in soda-lime glass. J.
Mater. Sci. 1983, 18 (9), 2785–2797.
8.61. Matthewson, M.J.; Kurkjian, C.R. Environmental effects on
the static fatigue of silica optical fiber. J. Am. Ceram. Soc.
1988, 71 (3), 177–183.
8.62. Krause, J.T. Zero stress strength reduction and transitions in
static fatigue of fused silica fiber lightguides. J. Non-Cryst.
Solids 1980, 38–39, 497–502.
8.63. Matthewson, M.J.; Rondinella, V.V.; Kurkjian, C.R. The
influence of solubility on the reliability of optical fiber. Optical
Materials Reliability and Testing, SPIE 1992, 1791, 52–60.
9.1 INTRODUCTION
The control of the chemical reactivity of ceramics with their
environment is one of the most important problems facing the
ceramics industry today. Through the study of corrosion
phenomena, one can learn best how to provide the control of
the chemical reactivity that will provide a maximum service
life expectancy at a minimum cost. Most methods used to
minimize corrosion have generally been methods that slow the
overall reaction rates. However, once a complete understanding
is available, one can attempt possibly to change the reaction
mechanism to something less harmful, in addition to slowing
373
Copyright © 2004 by Marcel Dekker, Inc.
374 Chapter 9
1. Heat transfer
2. Mass transfer
3. Diffusion-limited processes
4. Contact area
5. Mechanism
6. Surface-to-volume ratio
7. Temperature
8. Time
(9.1)
(9.2)
In the preservation of national monuments, engineers have tried
various materials to fill the porous limestones, marbles, etc. to
slow corrosion. Not only are the fillers used to eliminate open
porosity, but also they are used to consolidate or strengthen
friable portions of the structure. The most important parameter
of the fillers is that they too must be corrosion-resistant. Many
materials have been used to provide this filler/strengthening
characteristic from waxes to acrylic polymers to silicic ester-
type products [9.6]. According to Amoroso and Fassina [9.6],
these materials must have the following basic characteristics:
REFERENCES
9.1. McCauley, R.A. Evolution of flat glass furnace regenerators.
Glass Ind. 1978, 59 (10), 26–28, 34.
9.2. Brezny, R.; Semler, C.E. Oxidation and diffusion in selected
pitch-bonded magnesia refractories. J. Am. Ceram. Soc. 1984,
67 (7), 480–483.
9.3. Siljan, O.-J.; Rian, G.; Pettersen, D.T.; Solheim, A.; Schøning,
C. Refractories for molten aluminum contact Part I:
thermodynamics and kinetics. Refract. Appl. News 2002, 7
(6), 17–25.
9.4. Meadowcroft, D.B.; Cox, W.M. Dewpoint corrosion:
mechanisms and solutions. In Dewpoint Corrosion;
Holmes, D.R., Ed.; Ellis Horwood Ltd.: Chichester, UK,
1985. Chapter 2.
9.5. Harris, L.A.; Cross, D.R.; Gerstner, M.E. Corrosion
suppression on rutile anodes by high energy redox reactions.
J. Electrochem. Soc. 1977, 124 (6), 839–844.
9.6. Amoroso, G.G.; Fassina, V. Stone Decay and Conservation;
Elsevier: Amsterdam, 1983; 453 pp.
9.7. Lange, F.F.; Davis, B.I.; Metcalf, M.G. Strengthening of
polyphase Si3N4 materials through oxidation. J. Mater. Sci.
1983, 18 (5), 1497–1505.
9.8. Davies, G.B.; Holmes, T.M.; Gregory, O.J. Hot corrosion
behavior of coated covalent ceramics. Adv. Ceram. Mater.
1988, 3 (6), 542–547.
9.9. Gogotsi, Yu.G.; Lavrenko, V.A. Corrosion protection and
development of corrosion-resistant ceramics. Corrosion of
389
Copyright © 2004 by Marcel Dekker, Inc.
390 Glossary
393
Copyright © 2004 by Marcel Dekker, Inc.
394 Epilog